JP5401783B2 - Surface flaw detection method, detection apparatus and computer program for strip - Google Patents

Description

  The present invention relates to a surface wrinkle detecting device for a strip-like body such as a polymer sheet or a metal plate, and in particular, various types of surfaces of a steel strip or the like having different textures due to the material or surface treatment of the strip-like body such as a steel plate. The present invention relates to a surface wrinkle detection device and a detection method for a band-shaped body in which wrinkle inspection for detecting wrinkles is performed using a captured image.

  For example, in the manufacturing process of a steel plate or the like which is a kind of strip-shaped metal plate, surface defects (hereinafter simply referred to as defects) that may impair the quality of the product are discovered early in the manufacturing stage, and the manufacturing process or its It is necessary to prevent the occurrence of defects in subsequent products by improving the upper process or changing the manufacturing conditions. For this reason, the detection of wrinkles on the steel sheet moving in the production line is often performed. As one of the wrinkle inspection methods for detecting such wrinkles, a surface to be inspected such as a steel plate to be passed is illuminated with an illuminating device and is continuously photographed by an imaging device such as a CCD camera. There is an optical inspection method that detects wrinkles on the surface of a steel plate or the like based on image data.

  In the optical inspection method, a predetermined threshold value is set in advance for captured image data, and the detection of wrinkles is often performed based on whether or not the luminance level of the image data exceeds the threshold value. The following problems have arisen when capturing images of luminance levels.

  For example, in a general steel plate manufacturing process, various types of steel plates are manufactured, so that the appearance that appears on the surface of the steel plate has a large amount of reflected light from the surface of the steel plate, that is, surface brightness, for each steel plate material and surface treatment. Different. For this reason, a situation has arisen in which the luminance level of image data obtained by imaging the steel plate surface with an imaging device varies greatly for each steel plate. Further, even within a single steel plate (hereinafter also referred to as a steel plate coil), the surface brightness of each part of the steel plate varies depending on the position, and the brightness level of the imaged image data often changes. In addition, when a partially or wholly soiled steel sheet happens to pass through, a similar phenomenon occurs in which the surface brightness varies. Furthermore, since the surface of the steel sheet having a glossy mirror surface is highly dependent on the reflection angle, the surface luminance changes significantly depending on the surface properties, and the luminance level of the image data may change abruptly depending on the portion.

  Therefore, in order to obtain image data for wrinkle detection with an appropriate brightness level that is neither too bright nor too dark, the surface brightness is evaluated for each steel sheet material and surface treatment, and the steel sheets to be inspected are grouped. Therefore, it is necessary to perform so-called ground adjustment that adjusts the illumination level of the illumination device, the aperture level of the imaging camera that is the imaging device, and the like. As an apparatus for efficiently performing the formation adjustment, for example, Patent Document 1 sets and prepares a plurality of levels of imaging conditions that combine a plurality of illumination levels and aperture levels, and sets image data for each of the plurality of levels of imaging conditions. The combination of the illumination level and the aperture level is determined so that the average brightness value is a value around 128, which is the middle of the brightness range 0 to 255, for the brightness value in the specified range of the acquired image data. A data collection device is disclosed.

  Further, Patent Document 2 is provided with a plurality of amplifying units that amplify the image signal captured by one camera by 1 ×, 2 ×, 4 ×, and 8 ×, respectively. For image data, calculate the average brightness, select the image data with the average brightness closest to the predetermined reference brightness by using the switch, and output it to the image processing unit at the subsequent stage to handle changes in the brightness of the steel sheet. An apparatus is disclosed.

JP 2002-286650 A JP 2002-232778 A JP 2001-307067 A

However, the above prior art has the following problems.
In the technique disclosed in Patent Literature 1, image data is sampled from the imaging device for each of a plurality of levels that are combined by changing the illumination level and the diaphragm level sequentially for the steel plate to be inspected, and the optimal image data and By repeatedly determining the combination of the illumination level and the aperture level, the optimum illumination level and aperture level are found for different textures depending on the material and surface treatment of the steel plate. As a result, in this type of ground adjustment, it is necessary to prepare a preparation period in advance and collect image data for adjustment prior to collection of image data for wrinkle inspection. There was a problem that the inspection was not possible. Moreover, since it is necessary to carry out with respect to all combinations of the material and surface treatment of the steel plate to be inspected, there is a problem that a great deal of labor and time are required. In addition, there is a problem that regular ground adjustment is necessary because it is not possible to respond flexibly to changes in materials and surface treatments every time the steel plate manufacturing process is improved or new products are developed. It was.

  In the technique disclosed in Patent Document 2, a plurality of image data amplified at a plurality of magnifications is secured in advance for image data obtained from one imaging device, and the switch is performed when the brightness of the steel sheet changes. To select the image data with the optimal brightness. For this reason, it is necessary to accumulate a plurality of pieces of image data as compared with the case where processing is performed in one system, and there is a problem that the equipment cost increases. In addition, for the steel plates with a glossy mirror-like surface and a plurality of steel plates that have been subjected to various surface treatments, such as a steel plate with a rough matte surface, In order to apply a plurality of steel plates connected by welding to a line that passes continuously, dark image data obtained by an imaging device is amplified at a large magnification, and the obtained image data is S There are also problems that the / N ratio is poor, the dynamic range of the luminance level is small, and wrinkle detection must be performed with inappropriate image data.

  In view of the above problems, the object of the present invention is to provide a strip-like body such as a steel plate having a surface texture that varies greatly depending on the position, depending on the material and surface treatment of the steel plate, when the surface brightness is significantly different. When detecting wrinkles on the basis of image data obtained by imaging the surface, it is possible to obtain image data with an appropriate brightness level that is not too bright or too dark by suppressing the influence of fluctuations in the formation of the steel sheet. It is an object of the present invention to provide a surface wrinkle detection method and apparatus capable of accurately detecting harmful wrinkles with a simpler configuration than conventional ones.

The gist of the present invention for the above problems is as follows.
In the method for detecting surface wrinkles of a strip according to the present invention, the strip-shaped body in which the rear end portion and the front end portion of each of the plurality of strip-shaped bodies are connected by a connecting portion and the surface properties vary depending on the position is used as the material to be detected. A method for detecting surface wrinkles on a band-shaped body, which detects surface wrinkles on the surface of the band-shaped body from image data obtained by irradiating illumination light on the surface while moving the band-shaped body and imaging the irradiated portion with an imaging device An image data acquisition step of acquiring image data by continuously imaging the irradiated portion with the value of the imaging parameter recorded in the imaging parameter database in association with the type or lot of the band, and the image data a determination step of determining whether or not contain coupling part on the image data taken by a capture step, if it is determined not to contain connecting portion to the image data one frame of said image data A specified area image luminance calculating step for detecting a luminance value of a predetermined area for each image, and a luminance range set in advance for evaluating whether the detected luminance value of the specified area is suitable for flaw detection An image luminance determination step for determining whether the image is within or outside, and based on the determination result of the image luminance determination step, the value of the imaging parameter is maintained or changed and set by a predetermined change procedure An imaging parameter setting step, and a wrinkle detection step for detecting wrinkles on the surface of the band by image processing based on the image data, wherein the image data acquisition step is a new parameter set in the imaging parameter setting step. The subsequent image data is collected with the appropriate imaging parameter value, and the imaging parameter setting step changes the imaging parameter value. If the updated value of the imaging parameters to the imaging parameter database, it is determined to contain a coupling part on the image data taken by the image data acquiring step to come from the imaging parameter database, preset Based on the detected material information, the step of reading out the imaging parameter value associated with the type or lot of the band to be newly processed and setting it as a new value, the image data acquisition step includes The subsequent image data is collected with the value of the imaging parameter set as the new value .
The imaging parameter may be any one or a combination of exposure time, lens aperture, sensitivity gain, and illumination level.
In addition, the predetermined change procedure in the imaging parameter setting step is to set an increase / decrease step value ΔS of the imaging parameter in advance, and the luminance value is smaller than the luminance range set in advance by the determination result in the image luminance determination step. Sometimes, the value of the imaging parameter is increased by the increase / decrease step value ΔS, and when the luminance value is larger than the preset luminance range, the value of the imaging parameter may be decreased by the increase / decrease step value ΔS.
Moreover, the surface wrinkle detecting device of the present invention has a band-shaped body in which the properties of the surface fluctuate depending on the position, in which the rear end portion and the front end portion of each of the plurality of band-shaped bodies are connected by a connecting portion. An apparatus for detecting surface wrinkles of a band-like body that detects illumination on the surface of the band-like body from image data obtained by irradiating illumination light on the surface while moving the band-like body and imaging the irradiated portion with an imaging device. An imaging unit that continuously captures the irradiated portion and collects image data with an imaging parameter value recorded in an imaging parameter database in association with the type or lot of the strip, and the imaging unit A determination unit that determines whether or not a connection unit is included in the image data collected in step S, and when it is determined that the connection unit is not included in the image data, the image data is preset for each frame. Brightness of the selected area A specified area image luminance calculation unit for detecting the image, and whether the detected luminance value of the specified area is within or out of a predetermined luminance range for evaluating whether the image data is suitable for wrinkle detection An image luminance determining unit for determining the image luminance, an imaging parameter setting unit for maintaining or changing the value of the imaging parameter based on a determination result of the image luminance determining unit, or changing and setting in a predetermined change procedure, and the image data And a wrinkle detection unit that detects wrinkles on the surface of the belt-like body by image processing, and the imaging unit uses the new imaging parameter value set by the imaging parameter setting unit to store subsequent image data. collected, the imaging parameter setting unit, the updates the value of the imaging parameters to the scanning parameters database when it changes the value of the imaging parameters, taken by the imaging unit When it is determined that the connection portion is included in the image data, it is associated with the type or lot of the band to be newly processed from the imaging parameter database on the basis of preset material information to be detected. The image pickup parameter value is read and set as a new value, and the image pickup unit collects subsequent image data with the image pickup parameter value set as the new value. .
In addition, the predetermined change procedure in the imaging parameter setting unit is to set an increase / decrease step value ΔS of the imaging parameter in advance, and a luminance value is smaller than a luminance range set in advance as a result of determination by the image luminance determination unit Sometimes, the value of the imaging parameter is increased by the increase / decrease step value ΔS, and when the luminance value is larger than the preset luminance range, the value of the imaging parameter may be decreased by the increase / decrease step value ΔS.
Further, the computer program of the present invention uses, as a material to be detected, a belt-like body whose surface properties vary depending on the position, in which the rear end portion and the front end portion of each of the plurality of belt-like bodies are connected by a connecting portion. A computer program for causing a computer to execute processing for detecting wrinkles on the surface of a strip from the image data obtained by irradiating illumination light on the surface while moving and illuminating the irradiated part with an imaging device An image data acquisition process for capturing image data by continuously capturing the irradiated portion with the value of the imaging parameter recorded in the imaging parameter database in association with the type or lot of the band; a determining process of determining whether or not contain coupling part on the image data taken by the image data acquisition processing, the contains no connected portion to the image data determine The specified region image luminance calculation process for detecting the luminance value of a predetermined region for each frame of the image data, and whether the detected luminance value of the specified region is suitable for wrinkle detection Whether to maintain the value of the imaging parameter based on the determination result of the image luminance determination processing and the image luminance determination processing to determine whether it is within or out of the predetermined luminance range to evaluate Alternatively, the image data acquisition processing is performed by causing a computer to execute imaging parameter setting processing that is changed and set in a predetermined change procedure and wrinkle detection processing that detects wrinkles on the surface of the strip by image processing based on the image data. The subsequent image data is sampled with the new imaging parameter value set in the imaging parameter setting process, and the imaging parameter setting process is performed in the imaging parameter setting process. If the updated value of the imaging parameters to the imaging parameter database, it is determined to contain a coupling part on the image data taken by the image data acquisition process when changing the value of the meter, the imaging From the parameter database, based on the detected material information set in advance, it has a process of reading the value of the imaging parameter associated with the type or lot of the band to be newly processed, and setting as a new value, The image data acquisition process is a computer program that collects subsequent image data with the value of the imaging parameter set as the new value .

  According to the surface flaw detection method, detection apparatus, and computer program of the present invention, the imaging parameters for acquiring image data are flexibly corrected and changed based on the luminance value of a predetermined area of the captured image data. It becomes possible. And, when detecting wrinkles based on image data obtained by imaging the surface of a strip-like body such as a steel plate having a texture whose surface brightness is greatly different when irradiated with illumination light, depending on the material and surface treatment of the steel plate, It is possible to obtain image data having an appropriate luminance level that is not too bright or too dark without being affected by the formation. As a result, it is possible to obtain image data with appropriate brightness that is neither too bright nor too dark, and to stably obtain image data with appropriate brightness without adjusting the ground prior to detection of wrinkles. Therefore, harmful sputum can be detected accurately.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a diagram showing an outline of an embodiment of the apparatus for detecting surface wrinkles of a band according to the present invention. In this embodiment, the material to be detected, that is, the band to be inspected, is the steel plate 1 that flows through the manufacturing process of the steel plant.

  The said steel plate 1 shall be connected by welding by the welding point as a connection part the rear end and front-end | tip of each steel plate with which several steel plates are processed continuously. The steel plate 1 continuously transferred in the longitudinal direction by the roll 2 is strip-shaped in the width direction of the steel plate 1 by an illuminating device 3 (for example, using a fluorescent lamp or fiber illumination) disposed on the upper side (opposite side of the roll 2). The image is picked up by a CCD camera 4 composed of a line sensor, which is an image pickup device that is irradiated with light and arranged oppositely across the irradiation position on the surface of the steel plate 1.

  The CCD camera 4 is controlled by a camera controller 5 and continuously images the steel plate 1. The CCD camera 4 and the camera controller 5 constitute an imaging unit. Reflected images taken for each line of the irradiation position on the surface of the steel plate 1 are accumulated and configured as one frame image, and one piece of image data is interlocked with the movement of the steel plate 1 from the CCD camera 4. Then, it is output at a preset interval. There are a plurality of types of steel plates 1 that flow through the manufacturing process, and the width is about 1 m to 4 m. Therefore, in general, three or four CCD cameras 4 are arranged in parallel in accordance with the width of each steel plate 1. . The image data of the steel plate surface imaged and output by one CCD camera 4 is composed of, for example, 2048 × 512 pixels. The number of pixels of the image data may be set as appropriate depending on the size of the wrinkles to be detected or the desired resolution.

  In the present embodiment, the presence or absence of a welding point is detected by the welding point (or connecting portion) detection unit 6 for each image data captured with respect to a series of steel plates 1, and it is determined whether the adjacent steel plates have been switched. To do. When the image data is not switched, that is, when the image data is a single steel plate, the designated area image luminance calculation unit 7 calculates the luminance value of the area set in advance in the frame, and then the image luminance determination unit 8 sets in advance. It is determined whether or not it is within a certain luminance range. If the luminance value is outside this luminance range, the imaging parameter setting unit 9 issues a command to change the imaging parameters such as the aperture and shutter speed to the camera controller 5 to set the luminance range of the image data in advance. Adjust to enter.

  If the brightness value is within a preset brightness range, the imaging parameter database 10 is linked to information such as the material and surface treatment of the captured steel plate (hereinafter referred to as steel plate information), and the imaging parameter at this time is Store. The welding value detection unit 6, the designated area image luminance calculation unit 7, the image luminance determination unit 8, the imaging parameter setting unit 9, and the imaging parameter database 10 constitute a luminance value adjustment unit 14.

  In addition, when the welding point detection unit 6 determines that the steel sheet has been switched, the imaging parameter setting unit 9 immediately acquires the type (or lot) of the steel sheet from the imaging parameter database 10 based on preset steel plate information. ) Is read in association with the steel plate information, and a preset imaging parameter value is read out, and a command to change to the imaging parameter value is issued to the camera controller 5 to adjust the brightness of the image data.

  After that, the newly acquired image data is subjected to the above-described processing, and the wrinkle candidate extraction unit 11 in the wrinkle detection unit 15 extracts harmful wrinkle candidates. Then, the soot determining unit 12 determines the soot type and the degree of harmfulness, and displays the soot, particularly harmful soot, on the harmful soot display unit 13. Here, the eyelid candidate extraction unit 11 and the eyelid determination unit 12 may be configured as disclosed in Patent Document 3, for example. Alternatively, it may be configured by means for detecting wrinkles from other known image data.

  Next, the surface flaw detection method of the present invention will be described with reference to the flowchart shown in FIG. 2 in association with the function of each part of the surface flaw detection apparatus shown in FIG. In order to make the description simple and easy to understand, an example in which the exposure time of the CCD camera 4 is used as the imaging parameter for setting and changing the brightness of the obtained image data will be described for the imaging parameter. The imaging parameters that can be adjusted by changing the brightness of the image data when taking an image may be any of the lens aperture (iris) of the CCD camera 4, the gain of sensitivity, and the illuminance level of the illumination device. A combination of a plurality of combinations may be used, and may be selected as appropriate according to the operation method / function of the CCD camera 4 or the illumination device 3.

(Step S0)
The exposure time is set under the control of the camera controller 5, and the surface of the steel sheet is synchronized with a tracking signal output from the line speed detecting means (PLG) for detecting the moving speed of the steel sheet 1 by the CCD camera 4 at a predetermined rate. An image of the upper illumination part is taken, image data is collected, and input to the luminance value adjustment unit 14. The image data may be imaged without a gap on the surface of the steel plate 1 or may be imaged so that a part of each image data overlaps. (Image data acquisition step)

(Step S1)
In step S1, it is determined whether or not the image data includes a welding point (connection position of the steel plate coil). In order to determine whether or not the image data obtained by imaging a steel plate includes a welding point, image processing of the image data is performed, and the rear end and the front end of two steel plates to be processed in succession are welded positions or welds. When punching, a punch hole opened near the welding position may be binarized and extracted to determine the presence or absence of a welding point.

  In addition, usually, tracking is performed in which the sheet speed and position of the steel plate 1 are detected by the line speed detection means (PLG) of the steel plate 1. Since the timing of imaging with the CCD camera 4 for each production line is synchronized with the line passing speed, the image data is always taken with the same resolution, so the welding point is based on a tracking signal obtained separately from the production line. A signal may be obtained to determine that the image data includes a welding point (welding point (connecting portion) detection step).

  Here, if the image data being processed includes a welding point (step S1), the image data to be processed next may be an image of a steel material with a different product. Based on the steel plate information input in advance, the stored imaging parameters are read out and stored from the database 10 using the steel material information as a search key (step S8), and an exposure time change command is output to the camera controller 5 (step S4). , Imaging parameter setting step, data processing executed by imaging parameter setting unit).

  In addition, the manufacturing schedule which is the order which manufactures a steel plate is managed by the high-order process control (not shown) which manages a manufacturing line, and the kind and lot name of the steel plate 1 which are the detection material information, and This is realized by transmitting steel plate information including the material and surface treatment conditions to the imaging parameter setting unit 9 in advance. As a means for transmission, communication means such as TCP / IP (network: LAN) or RS232C may be used. In this embodiment, the surface flaw detection device includes a network I / O unit (not shown) for inputting steel material information via a LAN. Table 1 below is an example of a table of imaging parameters recorded in the imaging parameter database 10. As the steel material information of the steel plate 1 that passes through the production line, the combinations of the material and the type of surface treatment are arranged, and the exposure time is stored for that. Therefore, when the exposure time is read from the imaging parameter database 10, it is possible to immediately read from the steel material information transmitted from the host computer using the combination of the material and the type of surface treatment as a search key.

(Step S2)
In step S2, the brightness value I of the frame area preset in the frame is obtained for the image data obtained by imaging the surface of the steel plate 1. The said frame area | region should just be an inner side except the edge part (for example, less than 50 mm from the steel plate edge) of the steel plate about the plate width direction of the steel plate 1. In this way, when calculating the luminance value I, the influence of the luminance change of the steel plate edge portion and the luminance of the roll can be eliminated, and the luminance value I of the image data obtained from the steel plate can be accurately captured. Further, in order to reduce the load of data processing, the brightness is determined by setting a plurality of appropriately selected lines (for example, 2 to 10 lines) inside the edge in the width direction and appropriately selected in the longitudinal direction of the steel plate 1 as a predetermined area. May be obtained (designated area image brightness calculating step, data processing executed by the designated area image brightness calculating unit 7 in FIG. 1).

  Note that the luminance value I of the frame region may be used as a representative value by combining, for example, the maximum and minimum luminance values in the frame region in addition to the average luminance value, or other representative values may be used. Sometimes. Hereinafter, a case where an average value of luminance is mainly used will be described.

(Step S3)
If it is determined in step S1 that the image data contains no welding point, the image data being processed includes only the image data in the same steel plate. In step S3, it is determined whether or not the luminance value I of the image data calculated in step S2 satisfies a predetermined luminance range Id <I <Ib. Here, Id and Ib are set so that the image data is not too dark and not too bright in consideration of detection sensitivity, luminance S / N value, etc. so that wrinkle detection can be performed by appropriate image processing. Thus, when the image brightness is 0 to 255, it is preferable to set Id = 110 and Ib = 150. If the luminance value I of the image data satisfies this condition, it is determined that appropriate image data has been obtained, and the cocoon candidate extraction process is performed in step S5. Finally, in step S7, harmful poisoning is displayed. When the maximum value and the minimum value are used as the luminance value I, a luminance range may be set for each of the maximum value and the minimum value (image luminance determination step, processing executed by the image luminance determination unit 8).

(Step S9)
If the luminance value I of the image data does not satisfy the above condition (Id <I <Ib), in step S9, the imaging parameter is corrected so that the luminance value of the newly imaged image data satisfies the condition. . When the luminance value I is below the lower limit value Id, the image data is dark, so the exposure time is changed longer, and a command is sent to the camera controller 5 in step S4 to increase the exposure time of the CCD camera 4. If the value exceeds Ib, the image data is bright, so that the camera controller 5 is instructed in step S4 to shorten the exposure time of the CCD camera so that the exposure time is shortened (imaging parameter setting step, imaging parameter). Details of processing executed by setting unit 9).

  Specifically, the procedure for changing the exposure time, which is an imaging parameter, may be as follows. That is, when the increase / decrease step value ΔS of the imaging parameter is set in advance and the luminance value is smaller than the preset luminance range as a result of the determination by the image luminance determination step, the value of the imaging parameter is increased by the increase / decrease step value ΔS, When the luminance value is larger than the preset luminance range, the value of the imaging parameter is decreased by the increase / decrease step value ΔS.

  Next, a description will be given assuming that the process of obtaining the image data by calculating the luminance value I for the image data, changing the imaging parameter, and imaging again with the CCD camera 4 is obtained n times. The exposure parameter E is set in advance as an increase / decrease step value ΔS (%) of the exposure time E, the brightness value I of the nth image data is I (n), and the exposure set when the image data is captured. Assuming that the time is E (n), the exposure time E to be changed satisfies the following expressions (1) and (2) when I (n) <Id.

E = E (n) · (100 + ΔS) / 100 (1)
When I (n)> Ib,
E = E (n) · (100−ΔS) / 100 (2)

  For example, if the increase / decrease step value ΔS = 10%, the image data calculated luminance I (n) is bright and does not satisfy the condition when the exposure time E (n) = 150 ms set for the nth image data. In this case, the exposure time is changed to be shorter to 135 ms (= 150 × (100−10) / 100), and the image data calculation brightness I (n + 1) of the next n + 1-th obtained image is still bright. In this case, the exposure time is further changed to 121.5 ms (= 135 × (100−10) / 100). In the above description, the exposure time to be changed from the nth image data and the (n + 1) th image data is obtained. However, in the case of a steel plate in which the brightness of the image data is drastically changed, an appropriate interval period a is provided to provide the nth image data. The exposure time to be changed may be obtained from the data and the (n + a + 1) th image data.

  As another method, if the change slope of the image data calculated luminance ΔI (= I (n) −I (n + 1)) with the exposure time change amount ΔE is obtained, the exposure time E that satisfies the predetermined luminance range is immediately obtained. Since it can be calculated, it may be changed to the calculated exposure time E.

  Further, when the exposure time is changed in step S9, the imaging parameter setting unit 9 updates the changed new exposure time in the imaging parameter database 10. If the imaging parameter database 10 is updated every time an appropriate luminance range is satisfied in this way, the exposure time that satisfies the optimal luminance range can be achieved even when the luminance of the image data changes within the same steel plate. Since it is set at the time of subsequent imaging, it becomes possible to respond flexibly by updating the database even when the brightness of the captured image data changes from the conventional one by improving the manufacturing process or developing a new product (imaging Part of the parameter setting step).

  Here, FIG. 3 is a schematic example of a change in the luminance value I of the obtained image data when a series of time-series steel plates 1 are composed of the coil B, the coil A1, and the coil A2, and are manufactured in order. is there. The X axis in FIG. 3 corresponds to the time axis, and is a graph in which the number of frames of image data taken in order is plotted and the Y axis is a luminance value of the image data. Table 2 below exemplifies a part of the manufacturing management information of the coil B, the coil A1, and the coil A2 that is managed by the host computer.

  Here, the surface treatment A1 is a new product that has not been passed through the production line until now, and the default exposure time of 100 is stored in the imaging parameter database. When the coil B is switched to the coil A1 at the welding point, image data that is too bright without satisfying the predetermined luminance range Id <I <Ib is obtained. Therefore, in steps S9 and S4, the exposure time of the CCD camera 4 is changed, and the exposure time of the imaging parameter database 10 is updated to 150 when the luminance value I satisfies the predetermined luminance range. Thereafter, the coil A1 is subjected to wrinkle detection with image data having an appropriate image luminance, and when the coil is switched from the coil A1 to the coil A2 that is the same steel material information, the coil switching is detected in step S1, and in step S8. Since the exposure time 150 of the previous imaging parameter database 10 is read and inspected, soot detection is performed with image data having an appropriate image luminance that satisfies a predetermined luminance range Id <I <Ib.

(Steps S5 to S7)
In step S5, image quality improvement such as shading correction is performed on the image data, image analysis image processing such as labeling processing and geometric feature amount extraction is performed to extract the eyelid candidates, and in step 6, the extracted eyelids are extracted. A wrinkle determination process using a geometric feature amount or the like is performed from the candidate to determine the wrinkle type and the degree of harmfulness. Finally, in step S7, if there is a harmful flaw as a result of the wrinkle determination, it is displayed on the display device ( Wrinkle detection step, data processing executed by wrinkle detector 15).

  The surface flaw detection device according to the embodiment of the present invention described above includes a main storage device including a semiconductor memory, an external recording device such as an HDD and a DVD, an input / output device such as a keyboard and a mouse, and a display such as a liquid crystal display. Part of it can be constructed using a personal computer including the device. Moreover, it can also be configured as dedicated hardware combining a component having a function equivalent to those of these devices and a CPU. And it is realizable by operating the program produced so that the process of each said step of S0-S9 memorize | stored in RAM, ROM, etc. of the said personal computer might be run. Specifically, the program is recorded on a recording medium such as a CD-ROM, or provided to a computer via various transmission media. As a recording medium for recording the program, besides a CD-ROM, a flexible disk, a hard disk, a magnetic tape, a magneto-optical disk, a nonvolatile memory card, or the like can be used. On the other hand, as the transmission medium of the program, a communication medium (wired line such as an optical fiber, etc.) in a computer network (LAN, WAN such as the Internet, wireless communication network, etc.) system for propagating and supplying program information as a carrier wave A wireless line or the like.

  In addition, each function in the above-described embodiment is realized by executing a program supplied by the computer, and an OS (operating system) or other application software in which the program is operating in the computer. In the case where the functions of the above-described embodiment are realized in cooperation with the above-mentioned embodiment, or all or part of the processing of the supplied program is performed by a function expansion board or a function expansion unit of the computer. Such a program is also included in the present invention even when each function is realized.

  As described above, according to the surface flaw detection apparatus for a band according to the present embodiment, even a surface of a steel sheet having a texture whose surface brightness is greatly different depending on the material and surface treatment of the steel sheet is too bright or dark. Image data with appropriate brightness that is not too high can be obtained, and image data with appropriate brightness can be stably obtained without adjusting the ground prior to wrinkle detection. The remarkable effect of being able to do is exhibited.

It is a figure which shows the outline of one Embodiment of the surface flaw detection apparatus of this invention. It is a figure which shows the processing flow of embodiment of the surface flaw detection method of this invention corresponding to the surface flaw detection apparatus shown in FIG. It is a figure which shows an example of the graph which plotted the time transition of the luminance value of the steel plate image data connected with the welding point.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Steel plate 2 Roll 3 Illumination device 4 CCD camera 5 Camera controller 6 Welding point detection part 7 Specified area image luminance calculation part 8 Image luminance determination part 9 Imaging parameter setting part 10 Imaging parameter database 11 疵 candidate extraction part 12 疵 determination part 13 Harmful疵 Display unit 14 Brightness value adjustment unit 15 疵 Detection unit

Claims (6)

Illuminate the irradiated part of the surface while moving the strips, using the strips whose surface properties fluctuate depending on the position, with the rear end and front end of each of the strips being connected by a connecting part. A method for detecting surface wrinkles of a band-shaped body, which detects wrinkles on the surface of the band-shaped body from image data obtained by irradiating light and imaging an irradiated portion with an imaging device,
An image data acquisition step of capturing image data by continuously imaging the irradiated portion with the value of the imaging parameter recorded in the imaging parameter database in association with the type or lot of the strip,
A determination step of determining whether or not a connection portion is included in the image data collected in the image data acquisition step ;
A designated region image luminance calculation step for detecting a luminance value of a region set in advance for each frame of the image data when it is determined that no connection portion is included in the image data;
An image brightness determination step for determining whether the detected brightness value of the designated area is within or not within a brightness range set in advance for evaluating whether the image data is suitable for wrinkle detection;
An imaging parameter setting step for maintaining the value of the imaging parameter based on the determination result of the image luminance determination step or changing and setting in a predetermined change procedure;
A wrinkle detection step of detecting wrinkles on the surface of the belt-like body by image processing based on the image data,
The image data acquisition step collects subsequent image data with the new imaging parameter value set in the imaging parameter setting step,
The imaging parameter setting step updates the imaging parameter value to the imaging parameter database when the imaging parameter value is changed ,
When it is determined that the image data collected in the image data acquisition step includes a connecting portion, a band to be newly processed from the imaging parameter database based on preset material information to be detected It has a step of reading out the value of the imaging parameter associated with the type or lot and setting it as a new value, and the image data acquisition step is a value of the imaging parameter set as the new value, and thereafter A surface wrinkle detection method for a band-like body, characterized in that the image data is collected .   The method for detecting surface wrinkles of a strip according to claim 1, wherein the imaging parameter is any one or a combination of an exposure time, a lens aperture, a sensitivity gain, and an illumination level. The predetermined change procedure in the imaging parameter setting step is to set an increase / decrease step value ΔS of the imaging parameter in advance,
When the luminance value is smaller than a preset luminance range in the determination result of the image luminance determination step, the value of the imaging parameter is increased by the increase / decrease step value ΔS, and when the luminance value is larger than the preset luminance range, 3. The method for detecting surface wrinkles on a belt-like body according to claim 1, wherein the imaging parameter value is decreased by an increase / decrease step value ΔS. Illuminate the irradiated part of the surface while moving the strips, using the strips whose surface properties fluctuate depending on the position, with the rear end and front end of each of the strips being connected by a connecting part. A surface wrinkle detection device for a band-shaped body that detects wrinkles on the surface of the band-shaped body from image data obtained by irradiating light and imaging an irradiated portion with an imaging device,
An imaging unit that captures image data by continuously imaging the irradiated portion with the value of the imaging parameter recorded in the imaging parameter database in association with the type or lot of the strip,
A determination unit that determines whether or not a connection unit is included in the image data collected by the imaging unit ;
A designated region image luminance calculation unit that detects a luminance value of a region set in advance for each frame of the image data when it is determined that no connection unit is included in the image data;
An image luminance determination unit for determining whether the detected luminance value of the designated area is within a predetermined luminance range in order to evaluate whether the image data is suitable for wrinkle detection; or
Based on the determination result of the image brightness determination unit, the imaging parameter setting unit that maintains the value of the imaging parameter or changes and sets in a predetermined change procedure;
A wrinkle detection unit that detects wrinkles on the surface of the belt-like body by image processing based on the image data;
The imaging unit collects subsequent image data with a new imaging parameter value set by the imaging parameter setting unit,
The imaging parameter setting unit updates the imaging parameter value to the imaging parameter database when the imaging parameter value is changed ,
When it is determined that a connection part is included in the image data collected by the imaging unit, the type of the band to be newly processed from the imaging parameter database or based on preset detection target information or The image pickup parameter value associated with the lot is read and set as a new value, and the image pickup unit uses the image pickup parameter value set as the new value to store subsequent image data. An apparatus for detecting surface wrinkles of a band-shaped body, characterized by collecting . The predetermined change procedure in the imaging parameter setting unit sets an increase / decrease step value ΔS of the imaging parameter in advance,
When the luminance value is smaller than a preset luminance range as a result of determination by the image luminance determination unit, the value of the imaging parameter is increased by an increase / decrease step value ΔS, and when the luminance value is larger than a preset luminance range, 5. The apparatus for detecting surface wrinkles on a belt-like body according to claim 4 , wherein the value of the imaging parameter is decreased by an increase / decrease step value ΔS. Illuminate the irradiated part of the surface while moving the strips, using the strips whose surface properties fluctuate depending on the position, with the rear end and front end of each of the strips being connected by a connecting part. A computer program for causing a computer to execute processing for detecting wrinkles on the surface of a belt-like body from image data obtained by irradiating light and imaging an irradiated portion with an imaging device,
An image data acquisition process for capturing image data by continuously imaging the irradiated portion with the value of the imaging parameter recorded in the imaging parameter database in association with the type or lot of the strip,
A determination process for determining whether or not a connection part is included in the image data collected in the image data acquisition process ;
A designated area image luminance calculation process for detecting a luminance value of an area set in advance for each frame of the image data when it is determined that the image data does not include a connection unit;
An image brightness determination process for determining whether the detected brightness value of the designated area is within a brightness range set in advance for evaluating whether the image data is suitable for wrinkle detection; or
Based on the determination result of the image luminance determination process, the imaging parameter setting process for maintaining the value of the imaging parameter or changing and setting in a predetermined change procedure;
Causing the computer to perform wrinkle detection processing for detecting wrinkles on the surface of the belt-like body by image processing based on the image data;
The image data acquisition process is to collect subsequent image data with the new imaging parameter value set in the imaging parameter setting process,
The imaging parameter setting process updates the imaging parameter value to the imaging parameter database when the imaging parameter value is changed ,
When it is determined that a connection part is included in the image data collected in the image data acquisition process, from the imaging parameter database, a band to be newly processed based on preset detected material information It has a process of reading out the value of the imaging parameter associated with the type or lot and setting it as a new value, and the image data acquisition process is the value of the imaging parameter set as the new value, and thereafter A computer program that collects image data .

Images