JP4734176B2 - Wire abnormality detection method, wire abnormality detection device, and wire abnormality detection program - Google Patents

Description

  The present invention relates to a wire abnormality detection method, a wire abnormality detection device, and a wire abnormality detection program. More specifically, the present invention relates to an image abnormality detection method, an abnormality detection device, and an abnormality detection program based on image processing that can cope with an electric wire wound with an optical fiber without performing erroneous detection. Concerning conversion.

  For the maintenance and management of power transmission lines, inspections of power transmission lines by helicopters have been performed conventionally. One of the purposes of this inspection is the early detection of abnormal parts of electric wires damaged by lightning strikes. Early detection of abnormal parts of electric wires is indispensable for maintaining the reliability of power supply. FIG. 53 shows a processing flow of the electric wire inspection work system. The electric wire inspection work is performed by photographing the electric wire with a video camera mounted on the helicopter (S901), reproducing the video image (S902), and visually inspecting the video image by the inspector (S903). . The inspector visually confirms whether or not there is a broken wire that has broken a part of the electric wire or an arc mark (hereinafter also referred to as a weld mark) that may cause a broken wire in the future. In addition, in order to reduce oversight of an abnormal part of the electric wire, a check by slow reproduction of a video image and a double check by two inspectors are generally performed.

  However, the video image of the photographed electric wire is for a long time, and it takes an enormous amount of time for the inspector to visually confirm all of this and determine the abnormality. Moreover, it cannot be said that it is an efficient work considering that most of them are normal places. Moreover, since a great load is applied to the inspector, oversight or the like may occur. Moreover, since it takes time for visual confirmation, the cost is increased due to personnel costs and the like, which is not economical.

  Therefore, using the image processing technique for the captured video image, except for the part (frame image) that is determined to be normal, only the part (frame image) that is determined to be likely to be abnormal is left. A technique for reducing the burden on the inspector has been proposed by confirming an abnormal portion by visual inspection of the inspector only for the detected image (hereinafter referred to as an anomaly detection image).

For example, a technique has been proposed in which an image of a wire portion is cut out from an image of the wire taken and an abnormality of the wire is detected in the image cut out by image processing (Patent Document 1). By automatically detecting an abnormality in the shape of the wire, the labor of the inspector is reduced and the processing speed is increased.
JP 2005-57956 A

  As the demand for communication has increased in recent years, the expansion and maintenance of communication networks has become an urgent task, and problems such as insufficient communication capacity have also occurred in areas where optical fiber composite wires with built-in optical fibers are installed. ing. For this reason, in order to compensate for the shortage of communication capacity, an optical fiber line (hereinafter also referred to as an optical fiber) is newly wound around the existing optical fiber composite electric wire to increase the communication capacity and cope with the increase in communication demand. Many such wires exist and are expected to increase in the future. FIG. 20 shows an example of the image of the electric wire.

  However, when the technique of Patent Document 1 is applied to an electric wire (hereinafter also referred to as a wound optical fiber wire) in which an optical fiber is newly wound around the electric wire, the wound optical fiber is erroneously detected as an abnormal location, There is a risk of detecting almost all images as abnormal. Therefore, even if the conventional electric wire abnormality detection processing is performed on the electric wire, it is considered that the abnormality detection image becomes enormous and the burden on the inspector's visual inspection increases.

  In addition, in the prior art, the edge of the electric wire was detected by template matching that searches for a luminance pattern similar to the electric wire, but if there is a luminance pattern similar to the electric wire in the background, the detection of the edge of the electric wire is not successful. In some cases, erroneous detection may occur.

  In addition, in the prior art, since a static threshold is used for the color information value (mainly luminance value) to detect the abnormal part of the electric wire, depending on shooting conditions such as sunshine conditions and helicopter flight conditions, There was a case where a normal part was detected as an abnormal part.

  Conventionally, among the captured images of electric wires recorded on video tapes, DVDs, etc., the time period when the electric wires that do not require abnormality determination by image processing are not photographed and the electric wires that require abnormality determination by image processing are photographed. It is necessary to record the time code by the worker who took the image and determine the image to be processed based on the time code. The operation of recording the current time zone as a time code is complicated.

  Therefore, the present invention makes it possible to detect a wound optical fiber with high accuracy without detecting the wound optical fiber as an abnormal portion, and with respect to the electric wire. Thus, an object of the present invention is to provide an electric wire abnormality detecting method, an electric wire abnormality detecting device, and an electric wire abnormality detecting program capable of greatly reducing an abnormality detection image and reducing labor of an inspector. Furthermore, it aims at providing the electric wire abnormality detection apparatus and electric wire abnormality detection program which enable a high-speed abnormality detection process. It is another object of the present invention to provide an electric wire abnormality detection device and an electric wire abnormality detection program capable of improving the accuracy of detecting an edge of an electric wire and improving the detection accuracy of an abnormal part. It is another object of the present invention to provide an electric wire abnormality detection method, an electric wire abnormality detection device, and an electric wire abnormality detection program that make it possible to perform rapid processing by reducing the total amount of processing without requiring time code recording.

  In order to achieve this object, the wire abnormality detection method according to claim 1 includes a wire detection process for detecting, as a target image, a portion where a wire is photographed from an original image obtained by photographing a wire around which an optical fiber is wound. The representative value is obtained from the color information value of the pixel row sandwiched between two ideal outlines of the electric wire obtained from the target image among the pixel row on the vertical scanning line at each pixel position in the horizontal direction, and the representative value is the lowest. Searching for the left and right ends of the portion of the target image where the optical fiber line crosses around the horizontal pixel position that is the value, and crossing the horizontal pixel position between the left and right ends The other horizontal pixel positions are separated as non-crossing portions, and the horizontal pixel positions in the crossing portions are sequentially scanned as inspection points in the crossing portion, so that the representative value becomes the lowest value. In this inspection point, the inspection point where the representative value increases compared to the horizontal pixel position in front of the inspection point is the change point, and the inspection point after the representative value becomes the lowest value is the representative point. An inspection point whose value decreases in comparison with the horizontal pixel position in front of the inspection point is defined as a change point, and the change point is detected as an abnormal point. In a non-crossing portion, a representative value in the non-crossing portion is detected. An upper limit threshold and a lower limit threshold based on the average value are set in advance, and a wire abnormality detection process is performed in which a horizontal pixel position where the representative value exceeds the upper limit threshold or the lower limit threshold is detected as an abnormal location.

  According to another aspect of the present invention, there is provided an electric wire abnormality detection device that detects a portion where an electric wire is photographed as an object image from an original image obtained by photographing an electric wire wound with an optical fiber line, and stores the detected portion in a storage device. Information of a pixel column sandwiched between two ideal contour lines of an electric wire obtained from the target image out of a pixel column on the vertical scanning line at each pixel position in the horizontal direction by reading out the target image from the storage unit From the horizontal pixel position where the representative value is the lowest value in the center, search for the left and right ends of the part of the target image where the optical fiber line crosses, and A flag is set for each horizontal pixel position with the horizontal pixel position in between as a crossing portion and the other horizontal pixel positions as non-crossing portions. For the crossing portion, the horizontal pixel position in the crossing portion is sequentially scanned as an inspection point. At the inspection point before the representative value that is the lowest value, the representative value is the value of the inspection point. Further, it is determined whether or not the value is higher than the horizontal pixel position in front, and if the value is higher, the inspection point is detected as a change point, and the representative point is the representative point at the inspection point after the representative value becomes the lowest value. It is determined whether or not the value is lower than the horizontal pixel position in front of the inspection point. If the value is low, the inspection point is detected as a change point, and the target image from which the change point is detected is set as an abnormal image. Record and set the upper threshold and lower threshold based on the average value of the representative values in the non-crossing portion in advance for the non-crossing portion, and abnormal horizontal pixel positions where the representative value exceeds the upper or lower threshold Detected as a point, and the abnormal point is detected. In which and a wire abnormality detecting unit for recording target image as an abnormal image.

  The electric wire abnormality detection program according to claim 8 detects, as a target image, a portion where an electric wire is photographed from an original image obtained by photographing an electric wire wound with an optical fiber wire, and stores it in a storage device. Wire detection processing to be performed, and the target image is read from the storage device, and the pixel row sandwiched between the two ideal contour lines of the wire obtained from the target image among the pixel row on the vertical scanning line at each pixel position in the horizontal direction. Obtain a representative value from the color information value, and search for the left and right ends of the portion of the electric wire in the target image where the optical fiber line crosses around the pixel position in the horizontal direction where the representative value is the lowest value, A horizontal pixel position between the left and right edges is set as a crossing portion, and other horizontal pixel positions are set as non-crossing portions for each horizontal pixel position. It is determined whether it is a minute portion or a non-crossing portion, and for the crossing portion, the horizontal pixel position in the crossing portion is sequentially scanned as an inspection point, and the inspection point before the representative value becomes the lowest value. Then, it is determined whether or not the representative value is higher than the horizontal pixel position in front of the inspection point. If the representative value is higher, the inspection point is detected as a change point, and the representative value is the lowest value. In the inspection point after becoming, it is determined whether the representative value is lower than the horizontal pixel position in front of the inspection point, and if it is lower, the inspection point is detected as a change point, and the change point Is recorded as an abnormal image, and for the non-crossing portion, an upper threshold and a lower threshold are set in advance based on the average value of the representative values in the non-crossing portion, and the representative value is the upper threshold or the lower threshold. The pixel position in the horizontal direction exceeding Detecting, in which to execute the wire abnormality detection processing for recording target image the abnormal portion is detected as an abnormal image.

  Therefore, when detecting the abnormality of the electric wire wound with the optical fiber line, first, in the target image, the abnormality detection is divided into a transverse part where the optical fiber is photographed and a non-crossing part where the optical fiber is not photographed. Processing is performed. When separating the transverse part from the non-crossing part, the fact that the representative value is the lowest near the midpoint of the transverse part is used, and the horizontal part position of the horizontal part where the representative value is the lowest is the center. The left and right edges are searched for as the boundary between the crossing part and the non-crossing part. In addition, the detection of anomalies in the crossing section is a monotonic decrease or increase using the fact that the representative value in the crossing section decreases monotonously before the midpoint and increases monotonously after the midpoint. The pixel position in the horizontal direction that is not used is detected as the abnormal point as the change point. In addition, in the abnormality detection in the non-crossing part, the upper threshold and the lower threshold are set from the average value of the representative values in the non-crossing part, and the pixel position in the horizontal direction where the representative value exceeds the upper threshold or the lower threshold is set as the abnormal part. Detected.

  In addition, although the electric wire is wound around the whole electric wire in the wound type optical fiber wire, when the electric wire is photographed from one direction, on the photographed image, the portion where the optical fiber wire appears to cross ( There is a portion where the optical fiber line is hidden behind the electric wire and cannot be seen (non-crossing portion).

  According to a second aspect of the present invention, in the electric wire abnormality detection method according to the first aspect, the electric wire abnormality detection processing is performed in a horizontal direction where the representative value is the lowest value when separating the transverse portion and the non-crossing portion. From the pixel position, scanning is performed before and after the pixel position in the horizontal direction, and first, the left and right intersections where the representative value exceeds the average value of the representative values in the target image are searched. When the pixel at the midpoint in the vertical direction of the two ideal contours is the provisional left end and provisional right end, and the provisional left end and provisional right end are connected by a straight line, the intersection with the two ideal contours The pixel position in the horizontal direction where the corresponding pixel exists is set as the left and right ends.

  Therefore, with the boundary between the crossing portion and the non-crossing portion as the center, the horizontal pixel position where the representative value first exceeds the average value of the representative values, with the horizontal pixel position having the lowest representative value as the center. Further, when the pixels indicating the midpoint of the electric wire at the pixel position in the horizontal direction are connected with a straight line, the pixel position in the horizontal direction where the intersection with the two outlines of the electric wire exists is set.

  According to a third aspect of the present invention, in the electric wire abnormality detection method according to the first or second aspect, the electric wire abnormality detection process is set in advance before and after the inspection point as a center when detecting an abnormality in the crossing portion. The average value of the representative values at a predetermined number of horizontal pixel positions is compared with the average value of the representative values at a preset number of horizontal pixel positions before the inspection point, and Detection is performed.

  Therefore, when detecting the change point, the average values of the representative values at the plurality of pixel positions centering on the inspection point and the previous pixel position are compared.

  According to a fourth aspect of the present invention, in the electric wire abnormality detection method according to any one of the first to third aspects, the electric wire abnormality detection processing includes electric wires that form two ideal outlines of the electric wires from the target image. When determining the edge pixel of the candidate pixel, the absolute value of the difference value of the luminance value between two adjacent pixels in the vertical direction of the electric wire is equal to or greater than a preset threshold value, and a certain pixel area centered on the candidate pixel, When the variance of the difference between the luminance values is equal to or greater than a preset threshold value, the candidate pixel is set as an edge pixel.

  Therefore, when obtaining the edge pixel of the electric wire, the candidate pixel and a certain pixel centered on the candidate pixel are utilized by using the focal point that is focused on the electric wire portion with respect to the pixel that is a candidate for the edge pixel. The variance of the luminance value difference from the region (for example, 3 × 3 pixels) is obtained, and when the variance is large, the candidate pixel is set as an edge pixel.

  Further, the invention according to claim 5 is the electric wire abnormality detection method according to any one of claims 1 to 4, wherein the electric wire detection processing is performed with respect to a frame image in which the electric wire is not photographed in the original image. The abnormality detection process is not performed.

  Therefore, for the image determined to contain no electric wire in the electric wire detection processing, the subsequent processing is omitted, and the amount of data processed by the electric wire abnormality detection processing is reduced.

  According to a sixth aspect of the present invention, in the electric wire abnormality detection method according to any one of the first to fifth aspects, the abnormality of the electric wire is detected in advance when an abnormality is detected in a non-crossing portion. When the detection is performed with a certain periodicity in the range, the threshold is changed to a range wider than a preset upper threshold or lower limit threshold, and abnormality detection is performed again in the non-crossing portion.

  Therefore, when a plurality of abnormal locations are detected, it is determined whether the occurrence location of the abnormality has periodicity, and if it is determined to have periodicity, it may be a false detection of a normal location, The abnormality detection process is performed again after loosely resetting the threshold for abnormality detection in a wide range.

  As described above, according to the electric wire abnormality detection method according to claim 1, the electric wire abnormality detection device according to claim 7, and the electric wire abnormality detection program according to claim 8, the optical fiber wire wound around the electric wire is It is possible to detect an abnormality in the electric wire even with respect to the wound optical fiber line without erroneously detecting the damaged portion.

  Even if there is damage such as arc marks in the crossing portion around which the optical fiber line is wound, it can be detected as an abnormal location without being overlooked. For this reason, it is possible to increase the accuracy of the wire abnormality detection process.

  According to the electric wire abnormality detection method of the second aspect, it is possible to accurately separate the transverse portion around which the optical fiber is wound and the non-winding portion that is not wound in the electric wire. As a result, it is possible to improve the detection accuracy of each abnormal portion in the crossing portion and the non-crossing portion. Therefore, since it is possible to reduce the number of abnormality detection images, it is possible to reduce the burden on the inspector and speed up the abnormality detection work for the electric wires.

  In addition, according to the wire abnormality detection method according to claim 3, the influence of noise can be avoided by comparing the increase / decrease of the average values of the adjacent representative values, and the erroneous detection of the abnormal part can be reduced. Can do. In addition, the amount of calculation can be reduced, and the speed of wire abnormality detection processing can be increased.

  Moreover, according to the electric wire abnormality detection method of Claim 4, the detection of the edge pixel of the electric wire which forms the outline of an electric wire can be heightened. Thereby, since extraction of only the pixels constituting the electric wire from the target image can be performed with high accuracy and processing for detecting the abnormal portion of the electric wire can be performed, it is possible to detect the abnormal portion with high accuracy.

  Moreover, according to the electric wire abnormality detection method of the fifth aspect, it is possible to automatically exclude the frame image in which the electric wire is not photographed. As a result, the amount of processing for detecting the abnormality of the electric wire can be reduced, so that the entire processing can be speeded up. In addition, since the recording of the time code can be made unnecessary, the burden on the operator can be reduced and the work efficiency can be improved.

  In addition, according to the electric wire abnormality detection method of the sixth aspect, by dynamically selecting a threshold value suitable for the photographing environment, it is possible to reduce the erroneous detection of abnormal portions and to detect the electric wire abnormality with high accuracy. Thereby, since it is possible to reduce the number of abnormality detection images, it is possible to reduce the burden on the inspector and speed up the abnormality detection operation of the electric wire.

  Hereinafter, the configuration of the present invention will be described in detail based on embodiments shown in the drawings.

  In the electric wire abnormality detection method of the present invention, when detecting an abnormality of an electric wire wound with an optical fiber line, first, in the target image, it is divided into a transverse part where the optical fiber is photographed and a non-crossing part which is not photographed. Thus, different abnormality detection processing is performed for each. When separating the transverse part from the non-crossing part, the fact that the representative value is the lowest near the midpoint of the transverse part is used, and the horizontal part position of the horizontal part where the representative value is the lowest is the center. The left and right edges are searched for as the boundary between the crossing part and the non-crossing part. In addition, the detection of abnormalities in the crossing section is based on the fact that the representative value in the crossing section decreases monotonously before the midpoint and increases monotonously after the midpoint. The pixel position in the horizontal direction that is not used is detected as the abnormal point as the change point. In addition, in the abnormality detection in the non-crossing part, the upper threshold and the lower threshold are set from the average value of the representative values in the non-crossing part, and the pixel position in the horizontal direction where the representative value exceeds the upper threshold or the lower threshold is set as the abnormal part. Detected.

  First, the wire abnormality detection device of the present invention will be described. For example, as shown in FIG. 1, the wire abnormality detection device 1 includes an input interface 2 to which a video image of a wire is input, an output device 3 to which a captured image is output, and an input device such as a mouse and a keyboard. 4, a hard disk as an auxiliary storage device 5 in which data such as original images and target images are recorded, a RAM as a main storage device 6 in which temporary work data and the like are recorded, and a central processing unit (CPU) 7 etc. The above hardware resources are electrically connected through the bus 8, for example.

  The input interface 2 has a function for converting a signal input from the imaging means 9 such as a video camera or read from a medium such as a DVD or a video tape on which an image is recorded into data that can be processed by a computer. Each of the constituting frames has a function of recording in the auxiliary storage device 5 as image data. As such an input interface 2, for example, an existing NTSC-RGB converter, a frame grabber, an image capturing board for a personal computer, or the like may be used. The output device 3 is a display, for example, and displays an image in which an abnormality is detected, a user interface screen, and the like. The auxiliary storage device 5 stores the wire abnormality detection program of the present invention, and the computer functions as the wire abnormality detection device 1 by the CPU 7 reading and executing the program.

  Further, the electric wire abnormality detection device 1 detects, as a target image, an original image acquisition unit 10 that captures a captured image from the imaging unit 9 and records it as an original image, and a portion where an electric wire (hereinafter also referred to as a ground line) is captured. An electric wire detection unit 11 that performs electric wire detection processing and an edge of the electric wire (hereinafter also referred to as a contour line) is detected from the target image cut out by the electric wire detection unit 11 to detect an abnormality in the shape and color of the electric wire. The electric wire abnormality detection part 12 to be provided is provided. The original image acquisition unit 10, the electric wire detection unit 11, and the electric wire abnormality detection unit 12 can be configured by causing a computer to execute software executed by the CPU 7, and data necessary for the execution is loaded into the RAM 6. . In addition, the structure of the electric wire abnormality detection apparatus 1 is an example, and is not restricted to this.

  The electric wire abnormality detection method of the present invention will be described. In this wire abnormality detection method, as shown in the flowchart of FIG. 32, an image in which a wire is photographed is obtained from the imaging means 9, and an original image acquisition process (S100) that is used as an original image, and the region of the wire portion from the original image Wire detection processing (S200) for cutting out the wire, wire abnormality detection processing (S300) for detecting wire abnormality with respect to the cut-out target image, and an abnormal image for outputting the image in which the target image has not been cut out successfully or an abnormality detection image. It consists of output processing (S400).

  First, the original image acquisition process (S100) will be described. The original image acquisition process (S100) is a process of reading captured image data obtained by photographing an electric wire from a video camera or a medium on which video is recorded, and recording the image data in an auxiliary storage device 5 such as a hard disk. is there. The captured image data may be recorded in an external storage device provided separately from the electric wire abnormality detection device 1 and acquired through a LAN, the Internet network, or the like.

  In this embodiment, an aerial image of a wire taken by a video camera mounted on a helicopter is used as an image of the wire that is the inspection target. In this embodiment, the video camera generates an image of 30 frames per second, but the frame rate is not limited to this. In the original image acquisition process (S100), each frame image obtained from the video camera is converted into an RGB color model that can be processed by a computer. Converted to a grayscale image. Each frame image converted as described above is used as an original image in this embodiment. In the present embodiment, each pixel constituting the original image has a brightness value (luminance value) of 256 gradations from 0 (black) to 255 (white), for example, as a color information value. However, the color information value is not limited to the luminance value, and for example, a hue value may be used. The resolution of the original image is, for example, 720 pixels in the horizontal direction and 480 pixels in the vertical direction. However, the original image is not limited to a gray scale image and may be a color image, and the resolution is not limited to the above example.

  Hereinafter, in the present embodiment, the pixel position in the horizontal direction of the screen is referred to as a horizontal (direction) pixel number (for example, x = 1... 720), and the pixel position in the vertical direction of the screen is defined as a vertical (direction) pixel number ( For example, y = 1 ... 480). The horizontal direction and the vertical direction are the horizontal scanning line direction and the vertical scanning line direction on the screen, respectively. The electric wire is not necessarily photographed in the horizontal direction of the screen, and may be photographed in the vertical direction of the screen. Even in this case, the process may be executed with the longitudinal direction of the wire as the horizontal pixel number and the short direction of the wire as the vertical pixel number. In the present embodiment, the longitudinal direction of the electric wire is the horizontal direction of the screen, and the short direction of the electric wire is the vertical direction of the screen. In addition, the original image does not necessarily have to be photographed from the left end to the right end or from the upper end to the lower end of the image, and may include an image in which the electric wire is photographed only at the end of the image. Moreover, you may include the image of parts other than an electric wire, for example, a steel tower part.

  In the present embodiment, the original image data obtained by photographing the electric wire recorded in the auxiliary storage device 5 or the like by the original image acquisition process (S100) is converted into the subsequent electric wire detection process (S200) and the electric wire abnormality detection process (S300). By processing, the wire abnormality is detected. In this embodiment, while reading the original image data, the wire abnormality detection is performed by performing the wire detection process (S200) and the wire abnormality detection process (S300) for each frame of data of the original image. Is.

  An example of the original image acquisition process (S100) executed by the electric wire abnormality detection program of the present invention will be described along the flowchart of FIG. First, image data in which a wire is photographed is read from the video camera or a medium on which video is recorded via the input interface 2 (S101). Then, the image data recorded in the auxiliary storage device 5 is converted into, for example, an 8-bit grayscale image (S102). A Gaussian filter is applied to the original image thus obtained to remove noise components from the original image (S103). In order to avoid such a situation, sudden noise such as impulse noise appears in the image and a large peak is generated in the difference calculation in the edge detection process, causing erroneous detection of the edge. It is. This completes the original image acquisition process.

  Next, an electric wire detection process (S200) is performed. This is a process of cutting out a portion where an electric wire is photographed from an original image as a target image. Thereby, the data processing amount at the time of performing an electric wire abnormality detection process (S300) can be reduced significantly. That is, since the portion where the elongated electric wire is photographed is a very small part of the original image, even if the electric wire abnormality detection process (S300) is performed on the entire original image, it does not contribute to the edge detection accuracy, but is very large. Processing time is required. For this reason, first, a process of cutting out a portion where the electric wire is photographed from the original image is performed, the cut out region is stored as a target image, and a wire abnormality detection process (S300) is performed on the target image. The processing time is greatly shortened.

  An example of the electric wire detection process (S200) which the electric wire abnormality detection program of this invention performs is shown in the flowchart of FIG. In the electric wire detection process (S200) of the present embodiment, at least three search regions in which the horizontal range and the vertical range are predetermined are distributed in the horizontal pixel direction of the original image, and an image of the electric wire to be searched is used as a template. Is given in advance (S202), an area most similar to the template is searched in each search area (S206), and it is determined whether the search result for each search area satisfies a predetermined regularity. An image including at least a search result satisfying the above condition is cut out from the original image as a target image (S209).

  In this embodiment, the original image is captured at the TV rate (30 images / second). That is, a still image is captured every 33 milliseconds. Normally, it is considered that the position of the photographed electric wire in the image does not change significantly within 33 milliseconds, but the photographed image is taken from above the electric wire by a helicopter, so a gust of wind etc. Due to the influence of the above, there is an image in which the electric wire shooting position in the image is greatly moved and the electric wire is imaged only at the edge of the screen. For example, FIG. 2 shows an example of an image in which the electric wire 13 is photographed only at the edge of the screen. The original image includes an image in which the electric wire is not captured at the center of the image.

  The electric wire detection process (S200) will be described in detail. In this process, first, the frame counter n of the original image is set to 1 (S201). Next, a template registration process is performed (S202).

  In the template registration process (S202), for example, by specifying the pixels at the upper and lower positions of the electric wire displayed on the frame image, the intermediate position is recorded as the pixel position of the rectangular searcher 18a. Done. The searcher 18 is a window for searching for the position of the electric wire. Further, an area similar to the template is automatically searched in the vertical direction at a position separated by a predetermined number of pixels in the horizontal direction from the center of the searcher 18a, for example, at a position 30 pixels left and right. The three searchers 18a, 18b, and 18c serve as initial search results and record the respective vertical coordinate positions.

  A description will be given using an example of the user interface screen shown in FIGS. In the template registration process, the original image of the nth frame (usually the first frame) is displayed in the window 14 as shown in FIG. In this screen, the inspector selects the upper position 15 of the electric wire displayed in the window 14 and further selects the radio button “upper position of the electric wire” 16 with the input device 4 such as a mouse. Furthermore, if the selection is dragged to the lower side 17 of the electric wire (FIG. 4) and the selection is canceled (dropped), the rectangular searchers 18a, 18b, and 18c are automatically displayed in order from the left of the screen. (FIG. 5). Finally, the setting completion button 19 is pressed to complete the template registration process (S202). The coordinate position of the searcher registered in this way is stored in a storage device such as the RAM 6. If a template has already been registered, the above template registration process (S202) may be skipped.

  Next, when the n-th frame original image exists (S203; Yes), the original image is read (S204), and the midpoint of the wire detection position in the immediately previous frame (n-1 frame) is searched for. Three search areas are arranged as the center in the vertical direction (S205). When the original image is read up to the last frame (S203: No), the electric wire detection process (S200) ends. Since the wire search is performed for the original image of the first frame when the template is registered, 1 may be added to the frame counter n in the template registration process (S202). .

  The search area in the present embodiment is, for example, vertically long as shown in FIG. 6, the number X of pixels in the horizontal range is the same as the number of horizontal pixels in the template, and the number Y of pixels in the vertical range is the vertical pixels in the template. The number of pixels is larger than the number of pixels and equal to or less than the number of pixels in the vertical direction of the original image. Here, reference numeral 20 in FIGS. 6 and 7 indicates an original image, reference numeral 21 indicates a template, and reference numeral 22 indicates a search area. However, the size of the search area is not limited to the above example. Further, for example, as shown in FIG. 7, the search areas in the present embodiment are distributed at three places with a predetermined distance in the horizontal direction of the screen. However, the number of search areas may be three or more.

  Further, in the present embodiment, as shown in FIG. 8, the starting point 23 and the ending point 24 of the electric wire are obtained from the target image of the electric wire cut out from the original image, and then the middle point 25 is obtained from the obtained starting point 23 and ending point 24. The pixel position of the midpoint 25 is recorded and set as the center position of the search area for searching for the next frame. Specifically, when cutting out the electric wire (S209), a straight line connecting the positions of the three points in the search area is drawn to obtain two points that are the intersections of the straight line and the end of the image, and the midpoint 25 is We are going to calculate. As a result, even when the position of the electric wire is gradually deviated from the center of the image, the search area also moves with the movement of the position of the electric wire, so that the electric wire can be tracked. . By moving the electric wire search position for each image to be processed in this way, the electric wire can be tracked even when the electric wire position is gradually shifted from the center of the image.

  Next, a template matching process for searching for a region most similar to the template in the search region is performed (S206). Below, template collation processing (S206) is explained. For example, the degree of difference due to a square error is used to calculate the similarity between an area having the same size as the template in the search area (hereinafter referred to as a candidate area) and the template. In this case, if the brightness value of the template coordinates (x, y) is v (x, y) and the brightness value of the candidate area coordinates (x, y) is v ′ (x, y), the dissimilarity err is It is calculated by the following formula.

  A candidate area having the smallest difference err is set as an area most similar to the template in the search area, that is, a search result. However, the method of calculating the similarity between the candidate area and the template is not limited to the above example. For example, by using a method such as using cross-correlation for similarity calculation or determining similarity in the frequency domain, it is possible to perform more stable template matching with reduced influence of illuminance change.

  In order to further improve the accuracy of template collation, at the start of template collation processing, a portion having the largest difference in luminance values in the pixel row is obtained for each horizontal pixel number, and the template around the periphery of the pixel position is obtained. It is preferable to perform matching. This is because, in the original image, the electric wire portion is usually focused, and therefore, the luminance difference is considered to be greatest near the boundary between the electric wire and the background portion. By doing in this way, it becomes possible to improve the precision and processing speed of a template collation process.

  Next, wire movement position estimation (hereinafter also referred to as wire tracking) processing (S207), wire tracking success / failure processing is performed (S208), and an image including at least each search result is cut out from the original image. Is recorded in the auxiliary storage device 5 and the coordinate position of the midpoint 25 of the search result is recorded for the next search area arrangement (S210). Then, 1 is added to the frame counter n (S211), and the processing from S203 onward is repeated for the original image of the nth frame which is the next frame.

  FIG. 35 shows an example of a flowchart detailing the wire movement position estimation process (S207). In the movement position estimation process, first, the coordinate position of each search result in the horizontal direction of the screen is obtained (S207-1), and if the phase of the displacement of each coordinate position is aligned (S207-2; Yes), the movement position estimation process is performed. The process ends, and the process proceeds to a wire tracking success / failure process (S208). On the other hand, if the phase of the displacement of each coordinate position is not aligned (S207-2: No), the difference (movement vector) from the coordinate position of each search result in the immediately preceding frame is obtained (S207-3).

  The determination of the displacement phase (S207-2) will be described in detail below. In the present embodiment, as “regularity of search results” that is a criterion for determining success or failure of search, it is determined whether or not the phases of displacement of search results are aligned. The phase here refers to that used in “topology”. For example, if two graphics expressed by line segments have the same shape, it is determined that the two graphics have the same phase, that is, the phases are aligned. In the case of this embodiment, if the three search results are, for example, the first, second, and third search results in order from the left, the slope of the line segment connecting the first search result and the second search result, If the slopes of the line segments connecting the second search result and the third search result can be regarded as the same, it is determined that the phases are aligned. The reason why the success or failure of the search can be determined depending on whether or not the phases of the displacements of the search results are aligned will be described below. That is, the electric wires are not in contact with other objects except for the steel tower portion, and are in a state of hanging according to the physical laws. Therefore, the electric wire has a curved shape with a very small curvature, and can be regarded as a straight line within a small range within the screen without any problem. Therefore, the displacement of the wire position in the screen is relative to the movement of the helicopter or camera. The helicopter keeps flying at a constant speed along the electric wire and does not change its advancing direction suddenly. Also, the camera does not rotate, and only pans in the vertical and horizontal directions (moving the lens left and right while leaving the camera position unchanged). As described above, the electric wire can move in the vertical direction in the three search areas with time change, but the movement is usually aligned in the three search areas.

An example of numerically expressing whether or not the phase of displacement of each search result is aligned is shown below. _Assume that y _1i , y _2i , and y _3i are the vertical coordinate positions of predetermined points of each search result, for example, the center point in each search result. The subscript i = 0,..., P frame. Here, y ₁₀ , y ₂₀ , y ₃₀ are the coordinate positions in the current original image, y ₁₁ , y ₂₁ , y ₃₁ are the coordinate positions in the original image one frame before, and y _1p , y _2p , y _3p are The coordinate position in the original image before p frames is used. Then, based on the following formula, an average dy12rate of the difference between the distance between y _1i and y _2i and the distance between y _2i and y _3i in a certain period is calculated, and if the average dy12rate is within a predetermined constant value It is determined that the displacement phases of the search results are aligned, and when the predetermined values are exceeded, it is determined that the displacement phases of the search results are not aligned.

However, the determination of whether or not the phase of the displacement of each search result is aligned is not necessarily limited to that based on the above calculation. For example, the following more general calculation method may be used. For example, y1 = (y ₁₀ ,..., Y _1p ), y2 = (y ₂₀ ,..., Y _2p ), y3 = (y ₃₀ , y _3i using the above-described vertical coordinate positions y _1i , y _2i , y _3i . .., Y _3p ), and the cross-correlation of y1, y2, and y3 is calculated. Here, cor12 is a cross-correlation vector between y1 and y2, cor23 is a cross-correlation vector between y2 and y3, and cor31 is a cross-correlation vector between y3 and y1. In each cross correlation vector cor12, cor23, cor31, it is assumed that the maximum value is the j1, j2, j3th element, respectively. When these j1, j2, and j3 are both equal to or less than a certain value, it is determined that the phase of the displacement of each search result is aligned, and when the value exceeds the certain value, the phase of the displacement of each search result is not aligned. Judge. Note that j1, j2, and j3 are 0 when the phases are completely aligned. In the case of this calculation method, it is possible to detect whether or not the phase of displacement of each search result is aligned with higher accuracy. However, the calculation time is longer than the method based on Equation 2.

  Here, the electric wire tracking success / failure determination processing (S208) and the following processing may be executed for all frames of the original image. However, the original image does not include the electric wire, and an image obtained by photographing only the background, Includes an image of the part. This is an image of a steel tower portion or the like immediately after the start of shooting, just before the end of shooting or when the electric wire goes out of frame during shooting.

  In such a case, in the conventional technology, for example, the time code of the video is recorded for the time when the electric wire is shot in the video taken by the photographer, and the time code is specified by the image abnormality detection processing program. By doing so, it is set whether or not to be a target of processing. However, such work has been a burden for photographers. For this reason, in the electric wire abnormality detection method of the present invention, a frame removal process is performed in which an image frame in which no electric wire is photographed is automatically omitted without setting a time code or the like from the original image.

  Therefore, it is preferable to exclude images that do not include these electric wires from the following processing targets in advance. In the present embodiment, a frame removal process that does not perform subsequent processing is performed on an image that does not include an electric wire. As a result, the processing amount can be reduced to increase the overall processing speed, and erroneous detection can be prevented.

  The frame removal process will be described with reference to FIGS. (A) has shown the image which has succeeded in the tracking of an electric wire. (B) has shown the image which lost detection of the electric wire. If the electric wire is lost, each searcher 18 stops at the pixel position considered to be the electric wire most in the image. At this time, the determination is made based on whether or not the three searchers 18a, 18b, and 18c are in disorderly positions.

  In this embodiment, the disordered position is determined from the following two points. One is that each searcher 18 makes a determination based on a difference in pixel position from the immediately preceding frame. If the wire has been photographed and the tracking of the wire has been successful, the difference in the distance that each searcher 18 moves between two frames is small. In this embodiment, when the movement difference of each search element is equal to or greater than a preset threshold value, it is determined that the wire tracking has failed, that is, the frame has not been photographed. Further, when the difference between the distance (movement vector) between the searcher 18b and the searcher 18a and the distance (movement vector) between the searcher 18c and the searcher 18a is equal to or greater than a certain value, the wire tracking fails, that is, the wire Is determined to be a frame that has not been shot.

  In the present embodiment, a frame in which no electric wire is photographed is excluded from the subject of the electric wire abnormality detection process by satisfying one of the above two determination criteria, but the determination method is limited to this. It is not a thing. For example, any one of the above two determination criteria may be used. Further, for example, an angle formed by an intersection of a straight line connecting the searcher 18a and the searcher 18b and a straight line connecting the searcher 18a and the searcher 18c may be used as a criterion for determination. In this case, if the electric wire detection is successful, the line connecting the three searchers is aligned or approximated on a straight line, so the straight line connecting the searchers 18a and 18b and the straight line connecting the searchers 18a and 18c. The angle θ formed by the intersection of is small. However, for example, as shown in FIGS. 9C and 9D, when the electric wire is lost and a disordered position is detected as the electric wire position, the straight line connecting the searchers 18a and 18b and the searchers 18a and 18c are connected. The angle θ formed by the intersection of the straight lines increases. Therefore, when the angle θ is equal to or greater than a certain value, it may be determined that the wire tracking has failed, that is, the frame has not been photographed.

  As shown in FIG. 9E, when the wire tracking succeeds again, the subsequent wire tracking success / failure determination process (S208) is executed for the succeeding frames and thereafter. In the present embodiment, the removed frame image may be recorded in the auxiliary storage device 5 as a removed image separately from the abnormality detection image. In this case, it is also effective for the inspector to check the removed image.

  The above frame removal processing (S207-4 to S207-7) will be described with reference to the flowchart of FIG. In the frame removal process, first, it is determined whether or not there is any one that has moved by a predetermined pixel or more in the vertical direction compared to the immediately preceding frame among all the search elements (S207-4). If it is (S207-4: Yes), it is determined that there is no electric wire (S207-6). In the frame concerned, 1 is added to the frame counter without performing the subsequent processing, and the process returns to S203 (S207-7). ). In the present embodiment, it is determined that an image whose position has moved by 50 pixels or more is an image in which the electric wire is not photographed. However, the present invention is not limited to this.

If there is no searcher that has moved greatly (S207-4: No), then the searcher's movement vectors are compared (S207-5). In this embodiment, for example, the determination is made using Equation 3.
<Equation 3>
| (Movement vector of searcher 18b−Movement vector of searcher 18a) − (Movement vector of searcher 18c−Movement vector of searcher 18a) |> 10

  If Expression 3 is satisfied (S207-5: Yes), the three searchers exist at distant positions, so it is determined that the electric wire is not photographed (S207-6). Without performing the above process, 1 is added to the frame counter and the process returns to S203 (S207-7). Note that the determination criteria and the number of pixels serving as the determination criteria are merely examples, and are not limited thereto.

  Next, the wire tracking success / failure determination process (S208) will be described. In the technique of Patent Literature 1, when the phase of displacement of each search result is not aligned or when the similarity does not satisfy the defined standard, a template addition process is newly performed. In this case, the inspector actually needs to designate the electric wire portion in the original image on the user interface screen by using a pointing device such as a mouse, and the processing is temporarily stopped at the frame. On the other hand, according to the present invention, the template update process is automatically updated every fixed frame without the intervention of an inspector. Thereby, if the position of the electric wire is designated first, the electric wire abnormality detection process can be performed without performing the process involving the inspector such as re-registration of the template thereafter.

  In the present embodiment, only when the wire tracking is successful, a new wire template pattern is registered every certain number of frames. For example, the template pattern is newly updated every 100 frames. Two types of template patterns are used: an initial registration template registered in advance and an updating template for each fixed frame. Note that the number of registered template patterns is limited to two because there is a limitation on the capacity of the main storage device. However, the number of template patterns is not limited to this, and three or more types may be used.

  In the electric wire tracking success / failure determination process (S208) in this embodiment, it is determined that the electric wire tracking has failed if the inclination of the electric wire is equal to or greater than a preset threshold value, as compared with the image of the immediately preceding frame. Yes. The threshold value may be set arbitrarily and is not limited to the above example. In addition, the inclination of an electric wire calculates | requires the inclination of an electric wire from the coordinate position of the said starting point and an end point, calculating | requiring the starting point / end point of the electric wire in an image.

  Furthermore, in this embodiment, when it is determined that the tracking has failed because the inclination exceeds the threshold, the search area is expanded and the process returns to the template matching process (S206) to search for the wire again. For example, FIGS. 10A to 10C show examples of diagrams in which the search area is gradually expanded. In addition, the code | symbol 13a shows the electric wire in the present flame | frame, and 13b shows the electric wire position in the last flame | frame. For example, if the initial search area is 3 times the diameter of the wire and the search result in the 3 times search area is compared with the image of the immediately preceding frame, The search area is expanded to 7 times the diameter of the electric wire, and the process returns to the template matching process (S206) to search for the electric wire. Further, even if the search result is within the 7-fold search area, the search area is 15 times the diameter of the wire (normal captured image) if the inclination of the wire is tan θ is 0.1 or more compared to the image of the immediately preceding frame. The entire range in the vertical direction) and return to the template matching process (S206) to search for the electric wire. In other words, the search of the electric wire is performed by expanding the search area in three stages of 3 times, 7 times, and 15 times. It should be noted that how much the search area is expanded and how many times the search area is expanded can be arbitrarily set, and is not limited to the above example. In addition to setting a threshold value for the diameter of the electric wire, the threshold value may be set based on another value. For example, a threshold value may be set for a pixel such as 10 pixels or 20 pixels with the number of pixels as a threshold value.

  In the present embodiment, the reason why the search area is gradually expanded in a plurality of times is to limit the data processing range and perform faster processing. For example, if a threshold value of 15 times is set from the beginning, even if the position of the electric wire has not moved much, data processing in a 15 times range will be performed, resulting in increased processing waste. For this reason, the search area is gradually expanded, and only when the search area is not found, the search area is expanded. Thereby, it is possible to perform quick processing except for useless data processing. It should be noted that the search area may be expanded at least once, not necessarily twice, or three or more times.

  However, even when the search area is expanded in the entire vertical direction of the screen, there is a case where the electric wire cannot be searched. Naturally, this is the case when the wire itself is not photographed in the image, but it has been found that even when the wire is photographed, tracking of the wire may fail. This is due to changes in the brightness of the electric wires caused by the shooting environment such as sunlight, clouds and helicopter positions. In addition, since the video camera shoots the electric wire from above with a helicopter, the background image naturally becomes the ground background. In this case, for example, when the background (for example, a house, a factory, etc.) where the color information value is close to the electric wire is photographed, the detection of the electric wire may fail. That is, there is a case where erroneous detection is caused due to the shadow of the electric wire on the photographed image.

  Therefore, in the present embodiment, as described above, by storing the wire template pattern in the captured image in the main storage device in advance, the wire abnormality detection process is performed regardless of the shooting environment such as changes in sunlight. To be able to do. If the search of the electric wire fails even if the search area is enlarged, the electric wire is searched by changing the template pattern of the electric wire registered in advance from the image of the immediately preceding frame.

  In the present embodiment, as described above, the template pattern of the electric wire is input in advance at the start of the process, and the template pattern is stored and updated at a predetermined time interval arbitrarily specified during the process. I am going to do that. Specifically, if the search of the electric wire fails even if the above search area is expanded, the template pattern of the electric wire to be referenced is changed, and the normal search area is searched again. Yes. Furthermore, even if the search is performed by changing the template pattern of the electric wire to be referred to, if the search for the electric wire fails, the search area is further expanded to search for the electric wire. The template pattern of the electric wire may be updated at an arbitrary time interval. Alternatively, a plurality of electric wire template patterns may be stored, and the processing may be continued while changing the reference electric wire template pattern until the electric wire can be detected.

  From the results of the experiment, the process of enlarging the search area and determining the inclination of the electric wire, changing the template pattern of the electric wire to be referenced, and expanding the search area again and performing the process of determining the inclination of the electric wire, It has been found that wire detection is successful for all images. In the present embodiment, first, the electric wire is detected while changing the search area in three stages. If the electric wire cannot be detected even in that case, the template pattern of the electric wire to be referenced is changed, and then again. The electric wire is detected while changing the search area in three stages. The number of stages in which the process of expanding the search area and the process of changing the template pattern of the electric wire to be referred to can be arbitrarily set, and is not limited to the above example. The order in which the two processes are performed is not limited to the above example.

  As a result of the experiment, there has been almost no failure in tracking of the wire due to the above-described success / failure tracking processing of the wire, but there may be a case where it fails with a slight probability due to the influence of the background. For this reason, in this embodiment, if the wire search fails even if the above-described wire tracking success / failure determination process is performed, the image of the frame is recorded as a still image, and the process proceeds to the next frame. I have to. By doing in this way, even if it is a case where detection of an electric wire fails, it is supposed that a process will not be stopped. As a result, the inspector can only specify the position of the electric wire first, and thereafter can perform the process without any intervention until the end of the process, thereby enabling a quick process.

  The wire tracking success / failure determination process (S208) will be described with reference to the flowchart of FIG. If the inclination of the electric wire is equal to or greater than a preset threshold (for example, tan θ = 0.1) (S208-1; Yes), the search area is expanded from 3 to 7 times the diameter of the electric wire (S208-2). ) Template matching processing and wire movement position estimation processing are performed again (S206 to S207). Note that the number of times to expand the search area is not particularly limited, but here, the search area is expanded twice. Even if the search area is enlarged, if the inclination of the electric wire is equal to or more than the preset threshold value again (S208-3; Yes), the search area is enlarged from 7 times to 15 times the diameter of the electric wire (S208-4), Template matching processing and wire movement position estimation processing are performed again (S206 to S207). Even if the search area is expanded again, if the inclination of the electric wire is equal to or greater than a preset threshold value (S208-5; Yes), the template pattern of the electric wire to be referred to is changed, and the template matching process is performed again. An estimation process is performed (S206 to S207), and the electric wire tracking success / failure determination process is terminated. Moreover, if the inclination of the electric wire is less than a preset threshold at any time (S208-1, S208-3, S208-5; No), it is determined that the electric wire has been successfully tracked. The determination process ends. This completes the wire tracking success / failure determination process.

  When the process for all the original images is completed (S203; No), the electric wire detection process (S200) is ended, and the electric wire abnormality detection process is performed (S300). In the electric wire abnormality detection processing (S300), as shown in FIG. 37, when the electric wire is sound using the processing (S306) for detecting the edge pixel constituting the actual outline of the electric wire and the detected edge pixel. Processing for obtaining the ideal contour line (S307), and optical fiber crossing portion determination processing (S308) for separating the crossing portion around which the optical fiber is wound and the non-winding portion not wound in the target image, ideal Processing for detecting an abnormality in the color of the electric wire based on the luminance value of the region surrounded by the outline (S309, S310), and processing for detecting an abnormality in the shape of the electric wire based on the information about the edge pixel and the ideal outline ( S311).

  Edge pixels constituting an actual contour line of the electric wire are detected from the target image cut out from the original image and stored. Among the target images obtained by cutting out the electric wire portion from the original image, only the electric wire has the edge, that is, the contour information clearly appears, and the electric wire region can be accurately extracted by detecting the edge pixel.

  For example, with respect to an image as shown in FIG. 11, as shown by an arrow in the figure, a difference value of luminance values between two adjacent pixels is obtained in one direction in the vertical direction of the screen, for example, in the upward direction in the figure. FIG. 12 shows the relationship between the difference value and the vertical coordinate position. Among the points protruding from the difference value 0 appearing in FIG. 12, in other words, the difference value is larger or smaller than both of the difference values on both sides in the vertical direction of the screen (hereinafter also referred to as a peak). The thing with a large absolute value has shown the electric wire part. Therefore, if both ends of the peak indicating the electric wire portion can be detected, the electric wire region can be accurately extracted.

  Here, there is a method of searching for a peak where the absolute value of the difference value exceeds the threshold and determining the pixel constituting the peak as an edge pixel. However, in this case, an appropriate threshold fluctuates depending on the sun direction, weather, and the like. Therefore, accurate edge detection cannot be performed under a certain threshold. In addition, in the prior art, there is a problem that the electric wire portion cannot be accurately captured if there is a luminance pattern similar to the electric wire in the background.

  Therefore, in the electric wire abnormality detection method of the present invention, contour detection using the focal point of the electric wire is performed, and the edge detection is more accurate than the conventional method. The high accuracy of edge detection is indispensable for improving the accuracy of shape abnormality detection and color abnormality detection. This is because the shape of the electric wire cannot be accurately obtained unless the contour can be calculated accurately. Note that the focus of the wire here is uniform because the wire portion is clearly reflected if the wire is in focus and the background portion is blurred while the background portion is blurred. It means luminance value. In addition, even if the outline of the electric wire is 1 to 2 pixels, misjudgment affects the total luminance value and the like, and thus it is essential to improve the accuracy of edge detection.

  In the present embodiment, template collation processing (S206) is first performed to obtain a rough wire position, and the wire edge is searched in the vertical direction with the rough wire position as the center. This principle will be described with reference to FIG. 13 which is an enlargement of FIG.

  A point P1 where the absolute value of the difference value is maximum is searched for in FIG. Further, the peak P2 located on the outermost side is searched for around the maximum point P1.

  In the present embodiment, the electric wire peak P2 is searched in order from the outside of the electric wire with reference to P1. For example, in the case of P2 positioned on the lower side in the target image, the number of pixels (for example, 20 pixels) corresponding to half the diameter of the preset wire is subtracted from the vertical pixel number of P1, and further, the number of pixels set in advance (for example, 10 pixels). The search is started from the pixel position subtracting). In the search, the candidate pixel 26 of the peak P2 is sequentially searched from there toward the direction P1. Whether or not the pixel is a candidate pixel is determined based on whether or not the absolute value of the difference value is greater than or equal to a preset threshold value (for example, 5).

When searching for the candidate pixel 26 of the peak P2, as shown in Equation 4, the variance of the luminance value difference between the candidate pixel 26 and a certain pixel region 27 centered on the candidate pixel 26 is obtained, and the variance is greatly different. Then, the candidate pixel is determined as the peak P2 (see FIG. 14). Similarly, P2 located on the upper side in the target image is searched.
Note that v (i, j) indicates a luminance value at the pixel position.

This is because the above-mentioned focusing property, that is, when the wire is in focus, the wire portion is clearly reflected and includes various luminance values, while the background portion is blurred, so the brightness is uniform. It takes advantage of becoming a value. In the present embodiment, the pixel region 27 is set to 9 pixels of 3 × 3, and the variance difference threshold is set to σ ² = 9.0, but is not limited thereto.

  Two pixels constituting the peak P2 obtained in this way are defined as edge pixels. One of the two edge pixels is called a first edge pixel, and the other is called a second edge pixel. In the present embodiment, the upper edge pixel in the target image is the first edge pixel, and the lower edge pixel is the second edge pixel. The above processing is performed for all the horizontal pixel rows of the target image.

  Next, based on the edge pixel detected by the above-mentioned process, the process which calculates | requires the ideal outline in case an electric wire is healthy is performed. For example, as shown in the flowchart of FIG. 40, the low-reliability pixels that do not satisfy the predetermined standard are excluded from the first edge pixel group and the second edge pixel group, and the remaining first edge pixel group and second edge pixel group are Each is approximated by a straight line to obtain two ideal contour lines (S307-1 to S307-3).

  For example, in the present embodiment, the following processing is performed on the first edge pixel group and the second edge pixel group, respectively, so that low-reliability pixels that greatly impair the linearity are removed. That is, as shown in FIG. 41, each edge pixel is a pixel of interest, and the average of the coordinate positions in the screen vertical direction of edge pixels in a predetermined range in the horizontal direction of the screen around the pixel of interest, for example ± 15 pixels, is obtained. (S307-1-3, S307-1-5). Then, edge pixels whose vertical coordinate position is significantly different from the average, that is, edge pixels whose absolute value of the difference between the vertical coordinate position and the average is equal to or greater than a predetermined threshold value are excluded as low reliability pixels. (S307-1-4, S307-1-6).

  In addition, an edge pixel in which the distance between the paired first edge pixel and the second edge pixel is extremely narrow or wide, in other words, the vertical direction between the first edge pixel and the second edge pixel having the same horizontal coordinate position. Edge pixels whose direction coordinate position difference is not less than a predetermined upper limit value or less than a lower limit value are excluded as low reliability pixels (S307-1-7, S307-1-8).

  As a method of applying a straight line to the edge pixel group excluding the low reliability pixels, for example, a least square method is used in the present embodiment. However, the method of linear approximation is not limited to this example. The straight lines obtained based on the first edge pixel group and the second edge pixel group excluding the low reliability pixels are referred to as a first ideal contour line and a second ideal contour line, respectively.

  In this embodiment, the representative value in the vertical direction is obtained from the luminance value of the pixel column in the vertical direction of each screen sandwiched between the first and second ideal contour lines, and the representative value that deviates from the threshold value of the color information in advance in the horizontal direction of the screen. When a predetermined number of lines are continuous, it is determined that there is a possibility that an abnormality has occurred in the color of the wire due to a broken wire, an arc mark or a scratch. In the present embodiment, for each frame image, an average value of luminance values of pixel columns in the vertical direction sandwiched between the first and second ideal contour lines is obtained, and this average value is used as a representative value. The luminance value of an electric wire and the representative value which is the average value are demonstrated using FIG.

The lower left of the frame image is the origin, the vertical coordinate is the vertical pixel number, and the horizontal direction is the horizontal pixel number (see FIG. 19A). The average of the luminance values of the electric wires in the vertical direction existing on a certain horizontal pixel number (Xi) is set as a representative value Br (Xi) at the horizontal pixel number Xi and expressed by Equation 5 (see FIG. 19B).

Further, the average of the representative values Br (Xi) from the left end to the right end of the electric wires existing in the same frame image is μ, and the variance is σ ² . Note that μ and σ are obtained as shown in Equations 6 and 7. M is the number of pixels in the horizontal direction of the frame image.

  However, the method of obtaining the representative value is not limited to this example, and the mode value of the luminance values of the vertical pixel columns sandwiched between the first and second ideal contour lines may be used as the representative value in the vertical direction. . Note that the representative value Br (Xi) expressed by the mathematical expressions 5 to 7 is hereinafter referred to as a representative value Val (i).

  Next, an optical fiber crossing portion determination process (S308) is performed. As shown in FIG. 20, the portion of the frame image where the optical fiber line (indicated by reference numeral 53) is wound, that is, the portion 54 where the optical fiber line appears to cross on the image (hereinafter referred to as optical fiber). The crossing portion or the crossing portion) is not an object to be detected as an abnormality. However, since the optical fiber line is covered with a black cover, the brightness value of the transverse portion is the other part of the unwound wire, that is, on the image, the optical fiber line appears hidden behind the wire. The representative value is lower than that of the non-crossing portion 55 (hereinafter referred to as an optical fiber non-crossing portion or non-crossing portion) and is detected as an abnormal portion.

  Incidentally, when applying to the image which image | photographed the winding type optical fiber wire by the electric wire abnormality detection method of patent document 1, it confirmed that all the optical fiber wires were determined to be damaged. That is, it has been found that the technique described in Patent Document 1 cannot be applied to a wound optical fiber line. If an optical fiber line is detected as abnormal in this way, a considerable number of abnormalities will be detected, resulting in an increase in the number of abnormal detection images and an increase in the burden of visual inspection by workers. In other words, it is impossible to realize the object of the invention to speed up the process by discriminating only a portion having a high possibility of damage or the like by visual observation of an operator.

  Further, it has been found that the luminance value of the crossing portion of the optical fiber is minimized at the horizontal pixel number near the midpoint of the crossing portion. This is because the area (ratio) occupied by the optical fiber line in the vertical direction of the electric wire is the largest near the midpoint of the transverse part.

  Also, since the optical fiber wire is twisted and wound around the electric wire, the area occupied by the optical fiber on the electric wire in the frame image gradually increases as the optical fiber wire approaches the upper and lower outlines in the vertical direction of the electric wire. I found it to be smaller. This is because the area (ratio) occupied by the optical fiber line in the vertical direction of the electric wire decreases as the distance from the midpoint of the crossing portion increases.

  Due to the above properties, the representative value at the optical fiber crossing portion becomes the minimum value at the horizontal pixel number near the midpoint of the crossing portion, and further, the horizontal pixel number before that gradually approaches the minimum value, and the horizontal values thereafter. It was found that the pixel number gradually returned from the minimum value to the normal range where the normal optical fiber was not wound. FIG. 21 shows an example of a graph in which the relationship between each horizontal pixel number and the representative luminance value in the image of the wound optical fiber line shown in FIG. 20 is plotted. Reference numeral 56 indicates a luminance value curve in which representative values at each horizontal pixel number are plotted. As indicated by reference numeral 57, the change in the luminance value at the transverse portion is the minimum at the horizontal pixel number near the midpoint. The horizontal pixel numbers before and after that indicate monotonically decreasing and monotonically increasing.

  Based on the above knowledge, the wire abnormality detection method of the present invention uses the characteristics of the change in the luminance value in the wound optical fiber line to determine the crossing portion and the non-crossing portion of the optical fiber, It is possible to detect the crossing portion without erroneously detecting the optical fiber line as abnormal and without overlooking the damage. Further, it is possible to detect a non-crossing portion without overlooking damage.

  A method for separating the crossing portion and the non-crossing portion will be described with reference to an image diagram shown in FIG. FIG. 22 shows the electric wire 13 around which the optical fiber line 53 is wound, and the lower curve shows the change in the horizontal pixel number and the representative value in the wound optical fiber line. In the figure, reference numeral 56 indicates a luminance value curve in which representative values at respective horizontal pixel numbers are plotted, and a dotted line 58 indicates an average luminance value μ of the representative values in the frame. Reference numeral 42 denotes a first contour line, and 43 denotes a second contour line.

  First, the horizontal pixel number (hereinafter referred to as the minimum value) 59 of the portion having the minimum representative value is searched. This is because the minimum horizontal pixel number is considered to be the midpoint of the transverse portion of the optical fiber.

  In addition, each frame image of the video image obtained by photographing the wound optical fiber line usually includes the optical fiber line, but there is a frame where the optical fiber line is not photographed for some reason, that is, there is a crossing portion. There may be frames that do not.

  Therefore, in the present embodiment, it is considered that the minimum value of the representative value is not lowered for a frame image in which no crossing portion exists. Therefore, a threshold value (for example, luminance value 60) is set in advance as the minimum value, and the minimum value is set. The following processing is performed only when is below the threshold. When the minimum value exceeds the threshold value, the following processing is performed assuming that the optical fiber is not captured in the frame, that is, there is no crossing portion. This utilizes the fact that the minimum value of the representative value is lower in a frame in which a crossing portion exists than in a frame in which no crossing portion exists. Note that the determination method and threshold value of the frame image in which no crossing portion exists is an example, and the present invention is not limited to this.

  Next, two horizontal pixel numbers that are intersections 60 and 61 with the average value μ58 of the representative value in the frame are searched before and after the minimum value 59 as a center. That is, scanning is performed before and after the horizontal pixel number from the horizontal pixel number indicating the minimum value 59, and first, the horizontal pixel numbers having a representative value equal to or larger than μ are set as intersections 60 and 61. Further, the pixel at the midpoint in the vertical direction of the electric wires of the first ideal contour line 42 and the second ideal contour line 43 at the horizontal pixel numbers indicating the intersections 60 and 61 is a temporary left end 60a and a temporary right end 61a of the optical fiber. And

  Further, when the pixels indicating the temporary left and right ends 60a and 61a are connected by a straight line, a pixel which is an intersection of the first ideal contour line 42 and the second ideal contour line 43 is obtained, and this intersection point is determined as the left end of the optical fiber. 62a and right end 63a. The horizontal pixel numbers corresponding to the left end 62a and the right end 63a are indicated by 62 and 63, respectively.

  An example of the image obtained as described above is shown in FIG. In this embodiment, the left end 62a to the right end 63a of the optical fiber obtained by the above-described method are set as a transverse portion, and the other portion is set as a non-crossing portion. In addition, the dotted line 64 has shown the dotted line displayed on a corresponding location, when the electric wire abnormality detection program of this invention detects a crossing part.

  As described above, according to the optical fiber crossing portion determination process (S308) in the present embodiment, it is possible to accurately separate the crossing portion and the non-crossing portion of the optical fiber. Note that the method is not limited to the above-described method for separating the crossing portion and the non-crossing portion of the optical fiber. For example, although the determination accuracy is inferior, a predetermined number of horizontal pixel numbers before and after the minimum value are determined as the crossing portion. Other separation methods may be used.

  However, since damage such as arc marks can naturally occur even at the crossing portion, it is necessary to perform an abnormality detection process. In other words, if the crossing portion is simply discriminated from the non-crossing portion and the crossing portion is not detected abnormally, the damage occurring in the crossing portion is overlooked.

  Therefore, in the electric wire abnormality detection method of the present invention, different abnormality detection processes are performed on the crossing portion and the non-crossing portion in consideration of the difference in characteristics.

  The color abnormality detection process (S309) at the crossing portion is performed as follows. Whether the crossing portion is a non-crossing portion is determined by, for example, providing a determination flag H for each horizontal pixel number, setting H = 1 for a crossing portion, and setting H = 0 for a non-crossing portion. It is possible.

  Further, the process of the color abnormality detection process (S309) at the crossing portion and the color abnormality detection process (S310) at the non-crossing portion, which will be described below, does not matter. That is, the process of S309 is performed for the horizontal pixel number of the crossing portion, and the process of S310 is performed for the horizontal pixel number of the non-crossing portion.

  FIG. 24 (a) shows an example of a frame image in which a wound optical fiber wire including a broken wire is photographed. FIG. 24 (b) shows the horizontal pixel number in the image on the horizontal axis and the representative value on the vertical axis. The graph which plotted a certain luminance value is shown. FIG. 25 (a) shows another example of a frame image in which a wound optical fiber line including a broken piece of wire is photographed. FIG. 25 (b) shows the horizontal pixel number in the image, the vertical axis on the horizontal axis. The graph which plotted the luminance value which is a representative value on an axis | shaft is shown.

  Here, when there is no damage such as an arc mark in the crossing portion, for example, the luminance value changes monotonously as already shown in FIG. 21, whereas when there is damage, for example, FIG. As shown by reference numeral 65 in FIG. 25 (b), it has been found that the luminance value changes due to the influence of the damaged portion and there is a portion that does not change monotonously (hereinafter referred to as a change point). If there is no damage, it is considered that the number increased monotonously as indicated by a thick line portion 66 shown in FIGS. 24 (b) and 25 (b). In addition, the code | symbol 67 of Fig.24 (a) and Fig.25 (a) shows the applicable location of a strand break.

  For example, in the crossing portion of FIG. 25, the curve indicating the representative value bordering on the horizontal pixel number (minimum value) at which the luminance value is minimum is monotonically decreasing on the left and monotonically increasing on the right. However, the luminance value of the portion (symbol 65) corresponding to the damaged portion on the monotonously decreasing side increases without monotonously decreasing. Reference numeral 68 denotes a guide to be displayed when the abnormal wire detection program of the present invention detects an abnormal portion. In the present embodiment, the guide 68 is displayed above the horizontal pixel number corresponding to the abnormal part.

  In the color abnormality detection process (S309) at the crossing portion, damage detection at the crossing portion is performed using the distribution characteristic of the luminance value. For example, as shown in FIG. 26, when a luminance value curve 56 that monotonously decreases and increases monotonously before and after the minimum value 59 detects a change point 65 that does not change monotonously, an abnormality is detected at the crossing portion. Yes.

The change point 65 in the present embodiment is detected as follows, for example. An average of representative values of a predetermined number of horizontal pixel numbers (for example, five) is set to the left and right with a horizontal pixel number (hereinafter referred to as an inspection point) scanned in the transverse portion as the center. That is, the average μ _a of the representative values for the total 11 horizontal pixel numbers is obtained. Further, the average of the luminance values for five horizontal pixel numbers on the left and right is similarly taken for the point whose horizontal pixel number is 10 right from the inspection point. That is, the average μ _b of the representative values at 11 horizontal pixel numbers adjacent on the right side is obtained. Finally, the difference between the brightness values of μ _a and μ _b is calculated, and whether or not it is monotonically decreasing or monotonically increasing is set to a preset threshold, for example, when monotonic decreasing, the difference is 5 or more, and monotonic increasing Determines whether the difference is 5 or less. A point showing a change exceeding this value is determined as a change point, and is determined to be damaged.

  The reason for comparing the average of 11 pixel numbers in total when detecting the change point 65 is to avoid the influence of noise and to reduce the amount of calculation, and the number to be averaged is particularly limited. is not. The threshold value for determining the difference in luminance value for detecting the change point is set to 5 as a threshold value selected as a result of the experiment, and may be adjusted to an optimum value depending on the shooting environment or the like.

  Further, in the color abnormality detection process (S310) at the non-crossing portion, the threshold value of the color information to be compared with the representative value is calculated for each target image. For example, as shown in Formula 8, a value obtained by multiplying the standard deviation of the representative value obtained for a certain target image by α is added to the average value of the representative values obtained for the target image, and the value is added to the upper limit of the target image. And a value obtained by subtracting a value obtained by multiplying the standard deviation of the representative value by α from the average value of the representative values is set as the lower limit threshold value. Α is an arbitrary coefficient.

<Equation 8>
Threshold = average value of representative values (μ) ± coefficient (α) × standard deviation of representative values (σ)

  And, for example, when a representative value that is greater than or equal to the upper threshold is continuous for a preset number of pixels in the horizontal direction, or when a representative value that is less than or equal to the lower limit threshold is continued for a predetermined number of pixels in the horizontal direction It is determined that there is a possibility of wire abnormality such as wire breakage or arc scar. This utilizes the fact that when an electric wire has an arc mark or a flaw, the abnormal portion becomes darker or whitish than a normal portion of the electric wire.

  FIG. 27A shows a frame image in which a wound optical fiber wire is photographed, and FIG. 27B shows a graph in which the horizontal axis in the image is plotted on the horizontal axis and the luminance value is plotted on the vertical axis.

  In a portion that is a non-crossing portion of the optical fiber in FIG. Moreover, the part shown with the code | symbol 67 of FIG.27 (b) has shown that it can detect as the above, since the said strand cut part has exceeded the upper limit threshold value. Thus, it can be confirmed that the abnormality detection is possible by the color abnormality detection process (S310) in the non-crossing portion.

  Furthermore, in the electric wire inspection, even a slight abnormality is not allowed to be overlooked. Therefore, the threshold is set so that the detection sensitivity is strict so that a fine abnormality can be detected, that is, the coefficient α in Equation 8 is set small. Is assumed. However, by setting the threshold value narrow, it is possible to increase the accuracy of detecting an abnormal location, but conversely, a normal location may be detected as abnormal. It is conceivable that this may occur due to changes in the shooting environment such as the sunlight conditions and the position of the shadow of the helicopter even in the same single video.

  An example of the correspondence between the threshold and the detectable damage is shown in FIGS. FIGS. 28A and 28B are examples of broken wires. Α for detecting the damage 33 shown in FIG. 28A is 2.2, and α for detecting the damage 34 shown in FIG. 28B is 0.9. The damage shown in (a) is easily seen visually, and it is clear that immediate repair is necessary. However, the damage shown in (b) is difficult to detect visually, and the need for early repair is unknown. It is. In the conventional wire abnormality detection method, since α is a fixed value, for example, if α = 0.9, the detection sensitivity is more than double for the damage detection of (a), and excessive detection sensitivity. Therefore, there is a possibility that even a healthy thing may be judged as damaged.

  Therefore, it is preferable to set a plurality of threshold values in advance and dynamically change the threshold values according to the abnormality detection situation so as not to cause erroneous detection. The dynamic threshold setting process in this embodiment will be described in detail below.

  FIG. 29 is a graph in which a horizontal pixel number is plotted on the horizontal axis and a luminance value is plotted on the vertical axis for a frame image in which a portion that is not an abnormal portion is determined to be abnormal when the upper and lower thresholds are set statically. Show. The portion indicated by the reference numeral 35 exceeding the upper limit threshold is a normal portion, but is determined to be abnormal. In addition, the part shown with the code | symbol 36 was an arc trace. The threshold value α is set to 1.4.

  In this case, if the location indicated by reference numeral 35 is determined to be abnormal, it will lead to a large amount of erroneous detection, leading to an increase in the abnormality detected image. Therefore, it is necessary to set a threshold value so as not to cause such erroneous determination.

  However, if the value of the coefficient α is simply increased, on the contrary, a minute abnormal part cannot be detected abnormally and may be overlooked. Here, as shown in FIG. 29, it is found that there is a certain periodic change in the waveform in the case where an abnormal part is erroneously detected, and a threshold value is dynamically set based on the knowledge. For example, if there are more than one anomaly detection location and the appearance interval of the anomaly detection location is within a certain horizontal pixel interval range, it is determined that there is a periodic change, and a threshold value corresponding to the periodic change is determined. By changing the above, the abnormality detection accuracy has been successfully improved.

  FIG. 30 shows an image in which an electric wire with an arc mark is photographed. FIG. 31 shows the result of performing the color abnormality detection process on the image of FIG. In the figure, reference numeral 44 indicates a representative value in each horizontal pixel number, and reference numeral 45 indicates a lower limit threshold. FIG. 31 shows a state in which a representative value equal to or lower than the lower limit threshold appears continuously for 20 pixels or more, and it can be confirmed that an electric wire having an arc mark can be accurately detected by the color abnormality detection process.

  In addition, the electric wire abnormality detection process of the present invention can also accurately detect an abnormal portion called “laughter” or “kink” that is damage other than arc marks (see the examples).

  Further, in the color abnormality detection process in the non-crossing portion, for example, the average value of the luminance values of the pixel columns in the vertical direction of each screen sandwiched between the first edge pixel and the second edge pixel is set as the luminance value in the vertical direction. Is also possible. However, in this case, if there is a break in the wire part of the wire is jumping outside, the luminance value of the non-wire part, for example, the luminance value of the background is also included in the average value calculation, and an accurate color abnormality It cannot be detected. As in this embodiment, by calculating the average of the luminance values in the region sandwiched between the first and second ideal contour lines and using it as the representative value, the luminance value of the portion that is not the electric wire is included in the calculation of the average value. This provides an advantage of preventing and performing accurate color abnormality detection.

  In the shape abnormality detection process using the ideal outline (S311 in FIG. 37), a predetermined number of edge pixels that are separated from the ideal outline by a predetermined distance in the screen vertical direction continue in the horizontal direction of the screen. In such a case, it is determined that there is a possibility that the shape of the electric wire is abnormal due to a broken wire or the like. For example, when an edge pixel that is 10 pixels or more away from the ideal contour line appears over a length of 20 pixels, it is determined that there is a possibility of a wire abnormality such as a broken wire. This utilizes the fact that when a part of the electric wire is cut off and jumps to the outside, the edge pixels come together to some extent at some distance from the ideal contour line.

  FIG. 15 shows an image in which an electric wire causing a break in the strand is taken. FIG. 16 shows the result of performing the edge detection process on the image of FIG. In addition, a low-reliability pixel is identified by the above-described ideal contour estimation process from the edge pixels shown in FIG. 16, and the result of obtaining the ideal contour based on the remaining edge pixels excluding the low-reliability pixel is obtained. As shown in FIG. FIG. 18 shows the result of displaying the edge pixel of FIG. 16 and the ideal outline of FIG. 17 in an overlapping manner. In the figure, reference numeral 40 indicates a first edge pixel, reference numeral 41 indicates a second edge pixel, reference numeral 42 indicates a first ideal contour line, and reference numeral 43 indicates a second ideal contour line. FIG. 18 shows a state in which 20 pixels or more of edge pixels that are 10 pixels or more away from the ideal contour line appear continuously, and the wire having a broken wire is accurately detected by the above-described shape abnormality detection processing. I can confirm that I can do it.

  In the case of a wound type optical fiber line, there may be a case where the shape of the electric wire appears to swell due to the influence of the optical fiber line wound around the upper or lower part near the intersection with the crossing part in the non-crossing part. For the wound optical fiber line, it is preferable not to abnormally detect the bulge of the electric wire near the intersection existing in the non-crossing portion in the shape abnormality determination process (S311). For example, it is possible to prevent erroneous detection by not performing the shape abnormality determination process (S311) for the horizontal pixel numbers of the non-crossing portion for the preset number of pixels from the intersection of the crossing portion and the non-crossing portion. .

  In this case, the direction in which the optical fiber is wound indicates whether the optical fiber is along the upper side or the lower side of the electric wire. For example, the optical fiber crosses the electric wire obliquely upward to the right. If it is, the shape abnormality determination process is not performed only on the upper part of the preset number of pixels from the intersection of the right end of the crossing part and the non-crossing part, and the shape abnormality determination process is performed on the lower part. By doing so, the abnormality detection accuracy can be improved. In this case, the shape abnormality determination process is not performed only on the lower part of the number of pixels set in advance from the intersection of the left end of the transverse part and the non-crossing part, and the shape abnormality determination process is performed on the upper part. To do. If the winding direction of the optical fiber line is opposite, the part that performs the shape abnormality determination process may be reversed upside down.

  In addition, when there are a plurality of crossing parts, for example, two places at both ends of the frame image, false detection is performed by executing the above processing with the two parts as crossing parts and the remaining parts as non-crossing parts. Anomaly detection can be performed without any problem.

  In the present embodiment, it is assumed that a wound optical fiber line is captured in the captured image. Also, for example, it is possible to select whether or not the image includes a wound optical fiber line at the start of the wire abnormality detection process. If it is executed and does not include a wound optical fiber line, a conventional electric wire abnormality detection program may be executed. In addition, even if the wire abnormality detection process of the present invention is performed on an image obtained by photographing a normal electric wire other than the wound optical fiber line, the following abnormality detection process is processed as a non-crossing portion. It is possible to detect abnormality with high accuracy.

  In addition, the above-described optical fiber crossing portion determination processing and crossing portion color abnormality detection processing are not limited to optical fiber wires, and can be applied to conventional wires wound around electric wires.

  In addition, the optical fiber line usually has a luminance value lower than that of a normal electric wire part, but even if the luminance value is higher than that of the normal electric wire part, In the graph shown in FIG. 24B, the luminance value curve 56 is shown as a symmetric graph with respect to the average value 58 of the luminance values, so that adjustment of each threshold value of the color abnormality detection processing described above is performed. It is possible to cope with this.

  Here, the target image of the present embodiment is obtained from an aerial image, and an abnormal portion of the electric wire appears over several frames. Therefore, in the present embodiment, it is determined that an abnormality has occurred in the electric wire when the target image that is determined to have the possibility that the electric wire has an abnormality continues for a predetermined number, for example, two frames or more. I have to. Thereby, a more reliable electric wire abnormality detection can be performed. However, if the flight speed of the helicopter is slow, etc., abnormal parts appear in more continuous frames. Therefore, if more than a number of frames continue, it may be determined that an abnormality has occurred in the electric wire. good.

  When the electric wire abnormality detection process is completed, the abnormal image output process (S400) recorded in the log is performed. In the present embodiment, an original image in which the target image could not be cut out or a continuous image of the abnormally detected image is converted into a moving image file (for example, avi file, mpeg2 file, etc.) and can be reproduced on the output device 3 It is supposed to be. This is used as an abnormality detection image, and the inspector can check the moving image as it is, so that visual confirmation is easier. The inspector can visually determine the presence or absence of a problem by looking at the moving image displayed on the output device 3. For example, in the color abnormality detection process of the present embodiment, there is a possibility that it is determined that there is an abnormality even if there is a stain or the like on the electric wire, but the determination of whether it is a mere stain or an arc mark is finally inspected. It may be performed by a member. In addition, you may record as not only a moving image but a still image. For example, still images may be displayed in an album form. Moreover, both a moving image and a still image may be recorded, and an inspector may perform confirmation work using both images.

  Hereinafter, an example of the electric wire abnormality detection process (S300) which the electric wire abnormality detection program of this invention performs is demonstrated along the flowchart of FIGS. FIG. 37 shows a detailed flowchart of the electric wire abnormality detection process (S300). In this process, 1 is set in the image counter k (S301), 0 is set in the shape abnormality counter z (S302), and 0 is set in the color abnormality counter q (S303). If the k-th target image exists (S304; Yes), the target image is read (S305), and the edge detection process (S306), ideal contour estimation process (S307), and light are read for the read target image. Fiber crossing determination processing (S308), color abnormality detection processing (S309, S310), and shape abnormality detection processing (S311) are performed. After the completion of these processes, 1 is added to the image counter k (S312), and the processes from S304 onward are repeated for the kth target image corresponding to the next frame. If the above process is performed about all the target images (S304; No), an electric wire abnormality detection process will be complete | finished.

  FIG. 38 shows a detailed flowchart of the edge detection process. In this process, 1 is set to the screen horizontal pixel number i (S306-1). Then, it is determined whether the value of the horizontal pixel number i exceeds the horizontal pixel number i_max of the target image (S306-2). If the horizontal pixel number i is equal to or less than i_max (S306-2; Yes), 1 is set to the pixel number j in the vertical direction of the screen (S306-3). Then, it is determined whether the value of the vertical pixel number j exceeds the value obtained by subtracting 1 from the number of vertical pixels j_max of the target image (S306-4). If the vertical pixel number j is equal to or smaller than j_max−1 (S306-4; Yes), the difference between the luminance value of the pixel at the coordinate (i, j) and the luminance value of the pixel at the coordinate (i, j + 1) is obtained. The difference value (j) is recorded (S306-5). Then, 1 is added to the vertical pixel number j (S306-6), and the processes after S306-4 are repeated. When the value of the vertical pixel number j exceeds j_max-1 (S306-4; No), all the luminance value difference values between two adjacent pixels in the vertical pixel row at the horizontal pixel number i are obtained. Next, edge pixel search processing is performed (S306-7).

FIG. 39 shows a detailed flowchart of the edge pixel search process. In this process, the absolute value from among the recorded difference value difference value becomes maximum seek _{(j P1),} obtains the vertical pixel number _{j P1} corresponding to the difference value _{(j P1) (S306-7-1)} .

P2 located on the lower side of the target image is searched (S306-7-2 to 10). For example 20 pixels as half from _{j P1} of the wire diameter, the value obtained by subtracting another 10 pixels and j at the beginning (S306-7-2). Then, it is determined whether or not the difference value (j) corresponding to the vertical pixel number j can be a candidate pixel. For example, it is determined whether or not | difference value (j) |> 5 (S306-7-3). When the vertical pixel number j can be a candidate pixel (S306-7-3: Yes), the variance of the difference between the luminance value of the candidate pixel j and the luminance value of the 3 × 3 pixel area centered on j is calculated according to Equation 4. (Σ ² ) is obtained (S306-7-4). If σ is 3.0 or more (S306-7-5: Yes), the vertical pixel number j is set as the edge P2 (S306-7-10). On the other hand, when | difference value (j) |> 5 is not satisfied (S306-7-3: No) and when σ is less than 3.0 (S306-7-5: No), 1 is added to the vertical pixel number. (S306-7-6, S306-7-7), that is, processing is performed on the next pixel inside the electric wire. Here, it is determined whether or not the vertical pixel number j is within a threshold set in advance from P1, for example, 10 pixels (S306-7-8). When the edge P2 is searched in order from the outside pixel of the electric wire and becomes less than 10 pixels from P1 (S306-7-8: No), it is unlikely that there is an edge further inside. Therefore, it is determined that the edge detection has failed, and the edge P2 is tentatively set as the vertical pixel number (S306-7-9) obtained by subtracting half of the diameter of the wire from P1. On the other hand, when the vertical pixel number j is 10 pixels or more from P1 (S306-7-8: Yes), the process returns to S306-7-3 to determine whether the vertical pixel number can be an edge.

Next, P2 located on the upper side in the target image is searched (S306-7-11 to 19). Half from _{j P1} diameter of the wire at the beginning, further the value plus 10 pixels and j (S306-7-11). Then, it is determined whether or not the difference value (j) corresponding to the vertical pixel number j can be a candidate pixel. For example, it is determined whether or not | difference value (j) |> 5 (S306-7-12). When the vertical pixel number j can be a candidate pixel (S306-7-12: Yes), the difference between the luminance value of the candidate pixel j and the luminance value of the 3 × 3 pixel area centered on j is calculated according to Equation 4. (Σ ² ) is obtained (S306-7-13). If σ is 3.0 or more (S306-7-14: Yes), the vertical pixel number j is set as the edge P2 (S306-7-19). On the other hand, when | difference value (j) |> 5 is not satisfied (S306-7-12: No) and when σ is less than 3.0 (S306-7-14: No), 1 is subtracted from the vertical pixel number. (S306-7-15, S306-7-16), that is, processing is performed for the next pixel inside the electric wire. Here, it is determined whether or not the vertical pixel number j is within a threshold set in advance from P1, for example, 10 pixels (S306-7-17). When the edge P2 is searched in order from the pixel outside the electric wire and becomes less than 10 pixels from P1 (S306-7-17: No), it is unlikely that there is an edge further inside. Therefore, it is determined that the edge detection has failed, and the edge P2 is temporarily set to the vertical pixel number (S306-7-18) obtained by adding half the diameter of the wire from P1. On the other hand, when the vertical pixel number j is 10 pixels or more from P1 (S306-7-17: Yes), the process returns to S306-7-12 to determine whether the vertical pixel number can be an edge. Thus, the pixel difference value constitutes the farthest peak from the point j _P1 with the maximum, that is the edge pixel is obtained.

  When the first edge pixel and the second edge pixel are obtained for the horizontal pixel number i, the process returns to the process of FIG. 38, and 1 is added to the horizontal pixel number i (S306-8). The following process is repeated. As described above, the first edge pixel and the second edge pixel are obtained for each vertical pixel column constituting the target image, in other words, for each horizontal pixel number. When the value of the horizontal pixel number i exceeds the horizontal pixel number i_max of the target image (S306-2; No), the edge detection process in the target image is terminated.

  When the edge detection process is completed, an ideal contour estimation process (S307) is performed. FIG. 40 shows a flowchart detailing the ideal contour estimation process. In this process, first, a removal candidate pixel search process is performed (S307-1). Further, in this process, as shown in detail in FIG. 41, 1 is set to the screen horizontal direction pixel number i (S307-1-1). Then, it is determined whether the value of the horizontal pixel number i exceeds the horizontal pixel number i_max of the target image (S307-1-2). If the horizontal pixel number i is equal to or less than i_max (S307-1-2; Yes), the average value of the coordinate positions in the vertical direction of the screen is obtained for the first edge pixels existing at the horizontal pixel numbers i−Δi to i + Δi, and Ave1 (S307-1-3). Δi is set to 15, for example. Then, among the first edge pixels existing in the horizontal pixel numbers i−Δi to i + Δi, the edge pixels whose absolute value of the difference between the vertical coordinate position and the average value Ave1 is equal to or larger than a predetermined threshold value are removed. It records as a candidate pixel (S307-1-4). Similarly, for the second edge pixels existing at the horizontal pixel numbers i−Δi to i + Δi, the average value of the coordinate positions in the vertical direction is obtained as Ave2 (S307-1-5). Edge pixels in which the absolute value of the difference between the vertical coordinate position and the average value Ave2 is equal to or greater than a predetermined threshold are recorded as removal candidate pixels (S307-1-6). Further, the difference between the positions of the first edge pixel and the second edge pixel in the horizontal pixel number i in the vertical direction is obtained and set as ΔD (S307-1-7). When ΔD is greater than or equal to a predetermined upper limit value or less than a lower limit value, the first edge pixel and the second edge pixel at the horizontal pixel number i are recorded as removal candidate pixels (S307-1-8). Then, 1 is added to the horizontal pixel number i (S307-1-9), and the processing from S307-1-2 onward is repeated for the horizontal pixel number i. As described above, an edge pixel greatly deviating from the average of surrounding edge pixels whose vertical coordinate position is within a range of ± 15 pixels, or an edge where the distance between the paired first edge pixel and the second edge pixel is extremely narrow or wide Pixels are recorded as removal candidate pixels. When the value of the horizontal pixel number i exceeds the horizontal pixel number i_max of the target image (S307-1-2; No), the removal candidate pixel search process in the target image ends.

  When the removal candidate pixel is obtained, the process returns to the process of FIG. 40, and the remaining pixel excluding the removal candidate pixel, that is, the low-reliability pixel that impairs the linearity of the edge, is removed from the first edge pixel group by the least square method. A first contour line is created (S307-2). Similarly, a second contour line is created by the least square method using the remaining pixels obtained by removing the removal candidate pixels from the second edge pixel group (S307-3). Thereby, the ideal outline estimation process in the said target image is complete | finished.

  When the ideal contour estimation process ends, an optical fiber crossing portion determination process (S308) is performed. FIG. 42 shows a detailed flowchart of the optical fiber crossing portion determination process (S308). In this process, first, the transverse portion determination flag H is set for each horizontal pixel number (S308-1), and the screen horizontal pixel number i = 1 is set (S308-2). Then, it is determined whether the value of the horizontal pixel number i exceeds the horizontal pixel number i_max of the target image (S308-3). If the horizontal pixel number i is equal to or less than i_max (S308-3; Yes), the average value of the luminance values of the pixels located between the first contour line and the second contour line at the horizontal pixel number i is obtained, and the average The value is set as a representative value Val (i) (S308-4). Then, 1 is added to the horizontal pixel number i (S308-5), and the process for the horizontal pixel number i is repeated. Thereby, the representative value Val (i) is obtained for each horizontal pixel number i.

  Next, the horizontal pixel number that minimizes the representative value Val (i) is searched for and set to the minimum value Xmin, which is estimated as the midpoint of the crossing portion (S308-6). Next, an average value μ of the representative value Val (i) in the frame is obtained (S308-7), and the first value at the horizontal pixel number that is closest to Val (i) ≧ μ around the minimum value Xmin is obtained. The midpoints of the ideal contour line and the second ideal contour line are defined as the temporary left end and the right end, respectively (S308-8). Further, the temporary left and right ends are connected with a straight line to obtain an intersection point between the first ideal contour line and the second ideal contour line, and this intersection point is set as the left and right ends of the optical fiber (S308-9). In all horizontal pixel numbers from the horizontal pixel number to the rightmost horizontal pixel number, the crossing portion determination flag is set to H = 1 (S308-10). Thereby, the optical fiber crossing portion determination process in the target image ends.

FIG. 43 shows a detailed flowchart of the color abnormality detection process (S309) at the crossing portion of the optical fiber. In the color abnormality detection process at the optical fiber crossing portion, only the horizontal pixel number of the crossing portion determination flag H = 1 among the horizontal pixel numbers of the target image is targeted (S309-1). Next, the total counters Sum1 and Sum2 are each set to 0, and the counter L is set to -5 in order to average 5 pixels before and after the change point is detected (S309-2). Then, the average μ _a of representative values at a total of 11 horizontal pixel numbers centered on the inspection point and the average μ _b of representative values at a total of 11 horizontal pixel numbers centered on a point 10 to the right of the inspection point are obtained. In order to obtain the sum, while adding the representative value at each horizontal pixel number to the total counter (S309-3), until the counter L reaches 6 (S309-4), adding 1 to L (S309-5), the total Perform loop processing 11 times.

  When the loop processing from S309-3 to S309-5 is completed (S309-4: No), if the horizontal pixel number i is equal to or smaller than the horizontal pixel number of the minimum value Xmin, whether (Sum1-Sum2) / 11> 0 That is, whether or not it is monotonically decreasing, and if the horizontal pixel number i exceeds the horizontal pixel number of the minimum value Xmin, it is determined whether (Sum1−Sum2) / 11 <0, that is, whether or not it is monotonously increasing. (S309-6). If it is monotonous decrease and monotonous increase (S309-6: No), the process proceeds to S309-9. In contrast, when the horizontal pixel number i is equal to or smaller than the minimum horizontal pixel number, the difference from the sum of the right-side pixels has increased (it is +), and the horizontal pixel number i has the minimum horizontal value. If the pixel number is exceeded, but the difference is decreasing (-) (S309-6: Yes), it is determined whether the absolute value of the difference exceeds the difference threshold 5 (S309-7). ). When the difference threshold is exceeded (S309-7: Yes), 1 is added to A which is an abnormality detection number counter (S309-8), and the process proceeds to S309-9. If the difference threshold value is not exceeded (S309-7: No), it is not regarded as damage, and the process directly proceeds to S309-9.

  When the processing for all horizontal pixel numbers with the crossing determination flag H = 1 is completed (S309-9: Yes), when the abnormality detection number counter is not 0, the kth target image is recorded as a color abnormality candidate image. (S309-11). On the other hand, when there is an unprocessed horizontal determination flag H = 1 (S309-9: No), the process proceeds to the next horizontal pixel number (S309-10). Thus, the color abnormality detection process at the crossing portion in the target image is completed.

  Next, FIG. 44 shows a detailed flowchart of the color abnormality detection process (S310) in the non-crossing portion. In the color abnormality detection process in the non-crossing portion of the optical fiber, first, 0 is set to the counters u and u ′ and the abnormality detection number A (S310-1), and the obtained representative values Val (1) to Val (i_max) are obtained. Of these, the upper threshold value Val ′ and the lower threshold value Val ″ are obtained by Equation 8 using the horizontal pixel number with the crossing determination flag H = 0 (S310-2). Next, 1 is set to the horizontal pixel number i ( However, in the color abnormality detection process in the non-crossing portion, it is limited to the horizontal pixel number where the crossing determination flag H = 0, and it is determined whether the horizontal pixel number i is equal to or less than i_max (S310-4). If the horizontal pixel number i is less than or equal to i_max, it is determined whether the representative value Val (i) corresponding to the pixel number i is greater than or equal to the upper limit value Val ′ or less than or equal to the lower limit value Val ′ (S310-5). S310-7). If the value Val (i) is greater than or equal to the upper limit value Val ′, 1 is added to the counter u (S310-6). If the representative value Val (i) is less than or equal to the lower limit value Val ″, 1 is added to the counter u ′. Add (S310-8). Then, it is determined whether the value of the counter u or the counter u ′ is equal to or greater than a predetermined value u_max (S310-9). u_max is set to 20, for example. If both of the counters u and u 'do not satisfy u_max (S310-9; No), 1 is added to the horizontal pixel number i (S310-12), and the processing from S310-4 onward is repeated for the horizontal pixel number i. On the other hand, if the representative value Val (i) is not greater than or equal to the upper limit value Val ′ and not less than the lower limit value Val ″, 0 is set to the counters u and u ′ (S310-10), and 1 is added to the horizontal pixel number i. (S310-12), the process from S310-4 onward is repeated for the horizontal pixel number i.

  When at least one of the counters u and u ′ is greater than or equal to u_max (S310-9; Yes), a representative value Val (i) that is greater than or equal to the upper limit value Val ′ or less than or equal to the lower limit value Val ″ appears continuously for 20 pixels. Therefore, it is detected as an abnormal part, the abnormal part detection number A is set to A = A + 1 (S310-11), 1 is added to the horizontal pixel number i (S310-12), and S310-4 for the horizontal pixel number i. The following processing is repeated, at which time the counters u and u ′ reset 0 (S310-11).

  When the horizontal pixel number i becomes i_max (S310-4: No), it is determined whether or not an abnormality detection location exists (S310-13). When there is an abnormality detection location A (S310-13: Yes), dynamic threshold setting processing (S310-14) is performed. If there is no abnormality detection location (S310-13: No), the color abnormality counter q is reset to 0 (S310-18), and the color abnormality detection process for the target image is terminated.

  FIG. 45 shows an example of a flowchart of the dynamic threshold setting process (S310-14). First, it is determined whether or not more than a preset number (for example, three locations) of abnormality detection locations A have been detected (S310-14-1). If not detected, the process proceeds to S310-15. When three or more places are detected (S310-14-1: Yes), it is determined whether or not the appearance of the abnormal part has periodicity (S310-14-2). Whether or not it has periodicity is, for example, that the appearance interval of pixels corresponding to the abnormal location is within a preset periodic parameter (for example, a threshold in which 60 pixels have a certain width (for example, ± 5 pixels)) Judgment is based on whether or not When the abnormal part has periodicity (S310-14-2: Yes), the upper limit threshold and the lower limit threshold are relaxed (for example, coefficient α = α + 0.2) (S310-14-3), and after the change The process from S310-3 is executed again according to the threshold value α (S310-14-4). At this time, the abnormality detection number A is reset to zero. Note that the threshold value is changed only for the frame, and from the next frame, abnormality detection is performed with the original threshold value.

  Then, 1 is added to the color abnormality counter q (S310-15), and it is determined whether the counter q is equal to or greater than a predetermined value q_max (S310-16). For example, q_max is 2. If the color abnormality counter q does not satisfy q_max (S310-16; No), the color abnormality detection process for the target image is terminated. On the other hand, if the color abnormality counter q is equal to or greater than q_max (S310-16; Yes), the color abnormality candidate images are continuous for two frames or more, and therefore these color abnormality candidate images are recorded in the log (S310-17). ). Then, the color abnormality counter q is reset to 0 (S310-18), and the color abnormality detection process for the target image ends.

  46 and 47 show flowcharts detailing the shape abnormality detection processing. In this processing, 0 is set in the counters w and w '(S311-1 and S311-2), and 1 is set in the screen horizontal direction pixel number i (S311-3). Then, it is determined whether the value of the horizontal pixel number i exceeds the horizontal pixel number i_max of the target image (S311-4). If the horizontal pixel number i is equal to or less than i_max (S311-4; Yes), the difference in the coordinate position in the screen vertical direction between the first edge pixel at the horizontal pixel number i and the point on the first contour line at the horizontal pixel number i is determined. It calculates | requires and sets it as (DELTA) H1 (S311-5). Similarly, the difference between the coordinate positions in the vertical direction between the second edge pixel at the horizontal pixel number i and the point on the second contour line at the horizontal pixel number i is obtained and set as ΔH2 (S311-6). Then, it is determined whether the absolute value of ΔH1 and the absolute value of ΔH2 are equal to or greater than a predetermined value H_max (S311-7, S311-9). H_max is set to 10, for example. When ΔH1 is greater than or equal to H_max, 1 is added to the counter w (S311-8), and when ΔH2 is greater than or equal to H_max, 1 is added to the counter w ′ (S311-10). Then, it is determined whether the value of the counter w or the counter w ′ is equal to or larger than a predetermined value w_max (S311-11). For example, w_max is 20. If both of the counters w and w ′ do not satisfy w_max (S311-11; No), 1 is added to the horizontal pixel number i (S311-13), and the processing from S311-4 onward is repeated for the horizontal pixel number i. . On the other hand, if both ΔH1 and ΔH2 are less than H_max, 0 is set to the counters w and w ′ (S311-12), 1 is added to the horizontal pixel number i (S311-13), and the horizontal pixel number The process from S311-4 onward is repeated for i.

  When at least one of the counters w and w ′ is greater than or equal to w_max (S311-11; Yes), 20 edge pixels that are separated from the ideal contour line appear by 20 pixels in succession. Is recorded as a shape abnormality candidate image (S311-14). Then, 1 is added to the shape abnormality counter z (S311-15), and it is determined whether the counter z is equal to or greater than a predetermined value z_max (S311-16). z_max is set to 2, for example. If the shape abnormality counter z does not satisfy z_max (S311-16; No), the shape abnormality detection process for the target image is terminated. On the other hand, when the shape abnormality counter z is equal to or larger than z_max (S311-16; Yes), the shape abnormality candidate images are continuous for two frames or more, and therefore these shape abnormality candidate images are recorded in the log (S311-17). ). Then, the shape abnormality counter z is reset to 0 (S311-18), and the shape abnormality detection process for the target image ends. On the other hand, when the processing is performed until the value of the horizontal pixel number i exceeds the horizontal pixel number i_max of the target image (S311-4; No), 20 edge pixels that are separated from the ideal outline by 10 pixels or more are consecutive. In this case, the shape abnormality counter z is reset to 0 (S311-18), and the shape abnormality detection process for the target image is terminated.

  When the wire abnormality detection process for all frames is completed (S300), the abnormal image recorded in the log is output as a moving image file (abnormality detection image) (S400). Note that the process of outputting an abnormal image as a moving image file may be performed after completion of the process for all the frames, or may be performed in parallel during the process. Thus, the processing executed by the wire abnormality detection device 1 is completed by the wire abnormality detection program of the present embodiment.

  According to the present invention, it is not necessary for the inspector to visually check all the aerial images of the electric wires from the beginning, and it is only necessary to check some images and videos presented by the electric wire abnormality detection device 1. It is greatly reduced. Furthermore, by connecting the video camera to the electric wire abnormality detection device 1 and processing the image obtained from the video camera in real time, it is possible to perform detection of the abnormal portion of the electric wire at the site where the helicopter patrols, It becomes possible to reduce human resources and equipment resources related to work and to greatly reduce the overall cost.

  The above-described embodiment is an example of a preferred embodiment of the present invention, but is not limited thereto, and various modifications can be made without departing from the scope of the present invention. For example, in the above-described embodiment, each process is performed using a grayscale image as an original image, but each process may be performed using a color image as an original image. For example, when the original image is a grayscale image, the ridge line of the mountain may be determined as a search result in which the phase of displacement is aligned in the electric wire detection process. As an alternative, the hue value may be used in addition to the luminance value.

  Moreover, the processing procedure demonstrated using the flowchart about the electric wire abnormality detection program of this invention is an example, Comprising: It is not restricted to this. For example, various parameters can arbitrarily select optimum values according to the photographing environment, and there are of course places where the processing may be changed.

  Further, the edge detection process is not necessarily limited to the example of the above-described embodiment, and a conventional edge detection method may be adopted. In order to further increase the processing speed, a part or all of the processing shown in FIGS. 32 to 47 may be implemented as hardware. In particular, since the edge detection processing part is a general-purpose method, it has good compatibility with a commercially available image processing board and can be easily implemented as hardware. By using hardware, processing at a video rate becomes possible.

  Each parameter such as a threshold, for example, a range Δj for searching for a peak in the edge detection process, a reference H_max of a distance between an edge pixel for determining a shape abnormality and an ideal outline, a threshold coefficient α for determining a color abnormality, a continuous abnormality The criteria w_max and u_max for candidate pixels and the criteria z and q for consecutive abnormal candidate images may be adjusted as appropriate according to the shooting conditions.

  In the above-described embodiment, the wire abnormality detection process is performed on the target image cut out from the original image. However, in some cases, the wire abnormality detection process may be performed using the original image as it is as the target image. Moreover, you may use the target image obtained by the above-mentioned cutting-out process for the image processing for electric wire inspections other than a shape abnormality detection process and a color abnormality detection process.

  In addition, by using the color information value in addition to the luminance value, for example, the detection accuracy of the electric wire with a red line in which a red line for twist confirmation such as OPGW (optical fiber composite ground wire) is drawn You may make it improve. In addition, by correcting interlaced scanning (the capture time is different between the odd and even lines of the moving image to be processed), the sharpness of the image is improved and the abnormal part is sharpened to detect the abnormality. The accuracy may be improved.

Moreover, the electric wire abnormality detection method of this invention is applicable to linear rigid bodies other than an electric wire. For example, the present invention can be applied not only to electric wires but also to various cables and the like, and can be used for product inspection, for example.
Example 1

(Damage detection experiment for wound optical fiber)
The work amount reduction rate by the electric wire abnormality detection program of the present invention was 77%. This can be reduced by about 46 minutes in terms of one hour of tape.

  The work volume reduction rate of the wound optical fiber line was 54%. That is, since the inspector only has to detect the abnormality of the electric wire by merely visually inspecting the remaining 46% of the abnormality detection images, it can be confirmed that the inspection is effective for improving work efficiency and reducing costs.

In addition, when the damage was visually confirmed, all the damage was successfully detected, and no oversight occurred. From this experiment, it was confirmed that the wound optical fiber line can be detected by the wire abnormality detection program of the present invention without overlooking the damaged portion.
(Example 2)

(Damage detection experiment other than arc marks)
An experiment for detecting “laughter” and “kink”, which is damage other than wire breakage and arc marks, was performed by the electric wire detection abnormality detection program of the present invention.

  Electric wire laughter was detected by the electric wire detection abnormality detection program of the present invention. Note that laughter means a state in which the spacing between the strands is widened as indicated by reference numeral 51 in FIG. 48, and the strands can be broken if left unattended. Therefore, it is necessary to detect it as a form of damage.

  It has been found that the luminance value is higher in the portion where laughter exists than in the average luminance value of the pixels of the surrounding electric wires. FIG. 49 shows the relationship between the horizontal pixel number and the luminance value when laughter exists. It was confirmed that this can be detected with the same threshold value (for example, μ + 1σ) as that in the normal abnormality detection according to Equation 8.

  Next, the “kink” of the electric wire was detected. In addition, a kink means the state where the shape of the electric wire was distorted locally as shown to the code | symbol 52 of FIG. The kink is generated when the electric wire is installed while being twisted when laying the electric wire, and needs to be detected as one form of damage.

  When a kink exists, an actual contour exists inside the ideal contour line. That is, the representative value of the portion where the kink exists is a value obtained by adding the luminance value of the background to the luminance value of the actual electric wire. From the above, it has been found that the luminance value is lower in the portion where the kink exists than the average luminance value of the pixels of the surrounding electric wires. FIG. 51 shows the relationship between the horizontal pixel number and the luminance value when a kink exists.

  It was confirmed that “kink” can be detected with the same threshold (for example, μ−1σ) as that for detecting the arc mark.

  In addition, the kink not only causes distortion but also can swell as shown in FIG. In this case, it can be confirmed that the wire can be detected by the electric wire abnormality detection program of the present invention because it is considered to be the same as the case of a broken wire.

  In addition, the electric wire may be swollen due to corrosion. Also in this case, it was considered that the wire was broken, and it was confirmed that it could be detected by the electric wire abnormality detection program of the present invention.

  As described above, it has been confirmed that the electric wire abnormality detection program of the present invention can be detected without being overlooked as damage in the same manner as the arc mark without any particular change in the threshold value or the like. Thereby, it was confirmed that the electric wire abnormality detection program of the present invention can be detected without overlooking “laughter” and “kink”.

It is a schematic block diagram which shows an example of the electric wire abnormality detection apparatus of this invention. An image in which the electric wire is photographed only at the end portion of the photographed image is shown. It is an example of the user interface screen of a template registration process, and shows an example of the image in which the upper end of the electric wire was designated. It is an example of the user interface screen of a template registration process, and an example of the image from the upper end of the electric wire to the lower end is shown. It is an example of the user interface screen of a template registration process, and shows an example of an image on which a rectangular searcher is displayed. It is an example of the conceptual diagram explaining the principle which searches an electric wire in the electric wire abnormality detection method of this invention. It is another example of the conceptual diagram explaining the principle which searches an electric wire in the electric wire abnormality detection method of this invention. It is a conceptual diagram explaining the principle which searches the electric wire in the following flame | frame in the electric wire abnormality detection method of this invention. It is an example of an image in the frame removal process, (a) is an image when the tracking of the electric wire is successful, (b) is an image when the detection of the electric wire is lost, (c) and (d) are disordered An image when the position is detected as the electric wire position, (e) is an example of an image when the tracking of the electric wire succeeds again. It is an example of the conceptual diagram explaining the principle which searches an electric wire in the tracking success / failure determination process of the electric wire of this invention, (a) expands a search area 3 times, (b) enlarges a search area 7 times (C) is an example of a diagram showing a case where the search area is expanded 15 times. The image of the electric wire for demonstrating the principle of edge detection is shown. 12 is a graph showing a result of obtaining a difference value of luminance values between two adjacent pixels above the vertical direction indicated by an arrow in FIG. 11. The vertical axis of the graph indicates the vertical pixel number of the lower pixel for which the difference value is obtained, and the horizontal axis indicates the magnitude of the difference value. 13 is an enlarged view of a part of FIG. It is an example of the figure which shows a candidate pixel and pixel area for demonstrating the principle of edge detection. It is an image which shows an example of the electric wire which has raise | generated the strand break. It is the graph which displayed the edge pixel of the electric wire, a vertical axis shows a vertical pixel number, and a horizontal axis shows a horizontal pixel number. It is the graph which displayed the ideal outline of the electric wire, a vertical axis shows a vertical pixel number, and a horizontal axis shows a horizontal pixel number. It is a graph explaining the principle of the electric wire abnormality detection method of this invention, and displays the edge pixel of an electric wire and an ideal outline so as to overlap. The vertical axis of the graph indicates the vertical pixel number, and the horizontal axis indicates the horizontal pixel number. It is an example of the image of (a) electric wire for demonstrating a representative value and its average value, (b) It is a principle figure. It is an example of the image of a winding type | mold optical fiber line. It is the graph which showed the horizontal pixel number and luminance value in the image of FIG. The vertical axis of the graph indicates the luminance value, and the horizontal axis indicates the horizontal pixel number. It is an image figure for demonstrating the separation method of a crossing part and a non-crossing part. It is an example of the image by which the dotted line which displays a crossing part was displayed by the optical fiber crossing part determination process. (A) is an example of a frame image in which a wound optical fiber wire including a broken wire is photographed, (b) is a graph in which the horizontal axis number in the image is plotted on the horizontal axis and the luminance value is plotted on the vertical axis. It is an example. (A) is another example of a frame image in which a wound optical fiber wire including a broken wire is photographed. (B) is a plot of horizontal pixel numbers in the image on the horizontal axis and luminance values on the vertical axis. It is an example of a graph. It is a graph which shows an example of the luminance value curve in a crossing part, a horizontal axis shows a horizontal pixel number, and a vertical axis | shaft shows a luminance value. (A) is an example of a frame image obtained by photographing a wound optical fiber line, and (b) is an example of a graph in which the horizontal pixel number in the image is plotted on the horizontal axis and the luminance value is plotted on the vertical axis. It is a figure which shows an example of the abnormal location which can be detected with a threshold value, (a) is an example of the image of the electric wire containing the abnormal location which can be detected by (alpha) = 2.2. (B) is an example of the image of the electric wire containing the floor removal book detectable by α = 0.9. It is the graph which plotted the luminance value on the vertical axis | shaft and the horizontal pixel number on the horizontal axis about the frame image which determined the normal location as abnormal. The image which becomes the origin of Drawing 31 is shown, and the image of the electric wire with an arc mark is shown. It is a graph explaining the principle of the electric wire abnormality detection method of this invention, and represents the representative value and threshold value of the color information value of an electric wire. The vertical axis of the graph indicates the luminance value, and the horizontal axis indicates the horizontal pixel number. It is a flowchart which shows an example of the process performed by execution of the electric wire abnormality detection method and apparatus of this invention, and a program. FIG. 33 is a flowchart illustrating an example of a detailed process of the original image acquisition process of FIG. 32. FIG. It is a flowchart which shows an example of the process which detailed the electric wire detection process of FIG. It is a flowchart which shows an example of the process which detailed the movement position estimation process of the electric wire of FIG. It is a flowchart which shows an example of the process which detailed the tracking success / failure determination process of the electric wire of FIG. It is a flowchart which shows an example of the process which detailed the electric wire abnormality detection process of FIG. It is a flowchart which shows an example of the process which detailed the edge detection process of FIG. FIG. 39 is a flowchart showing an example of detailed processing of edge pixel search processing of FIG. 38. FIG. It is a flowchart which shows an example of the process which detailed the ideal outline estimation process of FIG. FIG. 41 is a flowchart illustrating an example of a process in which the removal candidate pixel search process of FIG. 40 is detailed. FIG. It is a flowchart which shows an example of the process which detailed the optical fiber crossing part determination process of FIG. It is a flowchart which shows an example of the process which detailed the color abnormality detection process (crossing part) of FIG. It is a flowchart which shows an example of the process which detailed the color abnormality detection process (non-crossing part) of FIG. It is a flowchart which shows an example of the process which detailed the dynamic threshold value setting process of FIG. It is a flowchart which shows an example of the process which detailed the shape abnormality detection process of FIG. 32, and shows the first half part of a process. It is a flowchart which shows the second half part of FIG. It is an example of the picked-up image of the electric wire in which laughter exists. 49 is a graph showing the relationship between horizontal pixel numbers and luminance values in the frame image of FIG. 48. It is an example of the picked-up image of the electric wire in which a kink exists. It is a graph which shows the relationship between the horizontal pixel number and luminance value in the frame image of FIG. It is another example of the picked-up image of the electric wire in which a kink exists. It is a figure which shows the processing flow of the conventional inspection work system of an electric wire.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Electric wire abnormality detection apparatus 10 Original image acquisition part 11 Electric wire detection part 12 Electric wire abnormality detection part 13 Electric wire 26 Candidate pixel 27 Pixel area 53 Optical fiber line

Claims (8)

  Among the pixel lines on the vertical scanning line at each pixel position in the horizontal direction, and an electric wire detection process for detecting a portion where the electric wire is photographed as an object image from an original image obtained by photographing an electric wire wound with an optical fiber line A representative value is obtained from the color information value of the pixel row sandwiched between two ideal outlines of the electric wire obtained from the target image, and the pixel position in the horizontal direction where the representative value is the lowest value is the center. Search the left and right ends of the portion of the electric wire in the target image where the optical fiber line crosses, and use the horizontal pixel position between the left and right ends as the horizontal portion, and the other horizontal pixel positions Are separated as non-crossing portions, and in the transverse portion, the horizontal pixel positions in the transverse portion are sequentially scanned as inspection points, and the inspection before the representative value becomes the lowest value. In the inspection point after the representative value becomes the lowest value, the inspection point where the representative value increases compared to the horizontal pixel position further before the inspection point is a change point. The inspection point at which the representative value decreases compared to the horizontal pixel position before the inspection point is defined as a change point, the change point is detected as an abnormal point, and the non-crossing portion An electric wire that presets an upper limit threshold and a lower limit threshold based on the average value of the representative values at a crossing portion and detects the horizontal pixel position where the representative value exceeds the upper limit threshold or the lower limit threshold as an abnormal location. An electric wire abnormality detection method by image processing characterized by performing abnormality detection processing.   In the electric wire abnormality detection process, when the transverse portion and the non-crossing portion are separated, scanning is performed from the horizontal pixel position where the representative value is the lowest to before and after the horizontal pixel position. First, the left and right intersections where the representative value exceeds the average value of the representative values in the target image are searched, and further, the midpoint in the vertical direction of the two ideal contour lines of the electric wire at the left and right intersections The pixel in the horizontal direction in which there is a pixel that is an intersection of the two ideal contour lines when the temporary left end and the temporary right end are connected by a straight line. The wire abnormality detection method according to claim 1, wherein the position is the left and right ends.   In the electric wire abnormality detection process, when an abnormality is detected in the crossing portion, an average value of representative values at a predetermined number of pixel positions in the horizontal direction around the inspection point and the inspection point 3. The wire abnormality according to claim 1, wherein the change point is detected by comparing with an average value of representative values at a preset number of pixel positions in the horizontal direction in front. Detection method.   When the wire abnormality detection process obtains edge pixels of the wire constituting the two ideal contour lines of the wire from the target image, a difference value of luminance values between two adjacent pixels in the vertical direction of the wire. When the variance of the difference in luminance value between a candidate pixel whose absolute value is equal to or greater than a preset threshold and a certain pixel area centered on the candidate pixel is equal to or greater than a preset threshold, the candidate pixel is The wire abnormality detection method according to claim 1, wherein the edge pixel is the edge pixel.   The electric wire abnormality processing according to any one of claims 1 to 4, wherein the electric wire detection processing does not perform the electric wire abnormality detection processing for a frame image in which the electric wires are not photographed in the original image. Detection method.   The wire abnormality detection process is wider than the preset upper limit threshold or the lower limit threshold when an abnormality is detected with a certain periodicity within a preset range when detecting an abnormality in the non-crossing portion. 6. The electric wire abnormality detection method according to claim 1, wherein the threshold value is changed to a range and abnormality detection is performed again at the non-crossing portion.   A portion where the electric wire is photographed is detected as a target image from an original image obtained by photographing an electric wire wound with an optical fiber wire, and the target image is read from the storage device and stored in a storage device. The representative value is obtained from the color information value of the pixel row sandwiched between two ideal outlines of the electric wire obtained from the target image in the pixel row on the vertical scanning line at each pixel position in the horizontal direction, and the representative value is obtained. Centering on the pixel position in the horizontal direction where the value is the lowest value, the left and right ends of the portion of the electric wire in the target image where the optical fiber line crosses are searched, and the horizontal position between the left and right ends A flag is set for each horizontal pixel position with the horizontal pixel position as a transverse part and the other horizontal pixel position as a non-transverse part. It is discriminated whether it is a non-crossing part, and for the transverse part, the horizontal pixel position in the transverse part is sequentially scanned as an inspection point, and the inspection point before the representative value becomes the lowest value. Then, it is determined whether the representative value is higher than the horizontal pixel position that is further in front of the inspection point. If the representative value is higher, the inspection point is detected as a change point, and the representative value is At the inspection point after the lowest value, it is determined whether the representative value is lower than the horizontal pixel position in front of the inspection point. Detected as a change point, the target image from which the change point has been detected is recorded as an abnormal image, and for the non-crossing portion, an upper threshold and a lower threshold based on the average value of the representative values in the non-crossing portion are set. Preset and the representative value is An electric wire comprising: an electric wire abnormality detecting unit that detects a pixel position in the horizontal direction that exceeds a limit threshold value or the lower limit threshold value as an abnormal location, and records the target image in which the abnormal location is detected as an abnormal image. Anomaly detection device. An electric wire detection process for detecting a portion where the electric wire is photographed as a target image from an original image obtained by photographing an electric wire wound with an optical fiber wire on a computer, and storing the target image in a storage device, and the target image as the storage device The representative value is obtained from the color information value of the pixel column sandwiched between two ideal contour lines of the electric wire obtained from the target image in the pixel column on the vertical scanning line at each pixel position in the horizontal direction. , Searching for the left and right ends of the portion of the electric wire in the target image where the optical fiber line crosses around the horizontal pixel position where the representative value is the lowest value, and between the left and right ends A flag that sets the horizontal pixel position as a transverse part and the other horizontal pixel position as a non-crossing part is set for each horizontal pixel position. It is discriminated whether it is a cut-off portion or the non-crossing portion, and for the transverse portion, the horizontal pixel positions in the transverse portion are sequentially scanned as inspection points, and the representative value is the lowest value. At the inspection point before this, it is determined whether the representative value is higher than the horizontal pixel position further before the inspection point, and if it is higher, the inspection point is detected as a change point. In the inspection point after the representative value becomes the lowest value, it is determined whether the representative value is lower than the horizontal pixel position in front of the inspection point. For example, the inspection point is detected as a change point, and the target image in which the change point is detected is recorded as an abnormal image. For the non-crossing portion, the average value of the representative values in the non-crossing portion is used as a reference. Preset upper and lower thresholds The horizontal pixel position where the representative value exceeds the upper limit threshold or the lower limit threshold is detected as an abnormal location, and an electric wire abnormality detection process for recording the target image where the abnormal location is detected as an abnormal image is executed. An electrical wire abnormality detection program.