JP2018004276A - Inspection device, inspection method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an inspection device, an inspection method and a program with which it is possible to inspect various defects with high accuracy.SOLUTION: The inspection device comprises: a read image acquisition unit 253 for acquiring a read image in which printed matter is read; a reference image acquisition unit 255 for acquiring a reference image that constitutes a printed matter inspection criterion; a difference image generation unit 257 for generating a difference image that indicates a difference between the read image and the reference image; a defect candidate area detection unit 261 for detecting on the difference image a defect candidate area in which is included a defect candidate portion that constitutes a candidate for defect; a feature quantity extraction unit 263 for extracting either the color feature quantity or shape feature quantity of the defect candidate portion; a strength calculation unit 265 for calculating a strength value that represents the visibility of the defect candidate portion on the basis of either the color feature quantity or the shape feature quantity; and an inspection unit 267 for inspecting on the basis of the strength value whether or not the defect candidate portion is a defect.SELECTED DRAWING: Figure 4

Description

  The present invention relates to an inspection apparatus, an inspection method, and a program.

  In printing that requires high quality such as production printing, quality inspection for printed matter is required. For example, an inspection device that inspects for the presence or absence of defects on a printed material by comparing a reference image generated from an original image of the printed material with a read image generated by electrically reading the printed material is known. Yes.

  Here, Patent Literature 1 detects the streak defect candidates present on the printed matter by the above-described method, and evaluates the visibility of the detected streak defect candidates to thereby detect the detected streak. A technique for accurately determining whether or not a candidate for a line defect is a line defect is disclosed.

  By the way, the defects detected from the printed matter by the inspection apparatus as described above include various types of defects such as point defects and line defects. However, the technique disclosed in Patent Document 1 described above is a line defect. It is only what evaluates visibility specializing in the streak defect which is a part of. Therefore, in the technique disclosed in Patent Document 1 described above, the visibility of the defects other than the streak defects is not evaluated, and it is not possible to inspect whether the defects are high-precision.

  The present invention has been made in view of the above circumstances, and an object thereof is to provide an inspection apparatus, an inspection method, and a program capable of inspecting various defects with high accuracy.

  In order to solve the above-described problems and achieve the object, an inspection apparatus according to an aspect of the present invention includes a read image acquisition unit that acquires a read image obtained by reading a printed matter, and a reference image that serves as an inspection reference for the printed matter. A reference image acquisition unit to be acquired, a difference image generation unit indicating a difference image between the read image and the reference image, and a defect candidate region including a defect candidate region that is a defect candidate is detected on the difference image. Based on at least one of the color feature quantity and the shape feature quantity, a defect candidate area detection section, a feature quantity extraction section that extracts at least one of the color feature quantity and the shape feature quantity of the defect candidate part, An intensity value calculation unit that calculates an intensity value that represents the visibility of the defect candidate site, and an inspection unit that inspects whether the defect candidate site is a defect based on the intensity value.

  According to the present invention, it is possible to inspect various defects with high accuracy.

FIG. 1 is a schematic diagram illustrating an example of a printed matter inspection system according to the present embodiment. FIG. 2 is a block diagram illustrating an example of a hardware configuration of the printing apparatus according to the present embodiment. FIG. 3 is a block diagram illustrating an example of a hardware configuration of the printed matter inspection apparatus according to the present embodiment. FIG. 4 is a block diagram illustrating an example of functional configurations of the printing apparatus and the printed matter inspection apparatus according to the present embodiment. FIG. 5 is a diagram illustrating an example of threshold information for defect candidate detection according to the present embodiment. FIG. 6 is a flowchart illustrating an example of a flow of a process performed by the printed matter inspection apparatus according to the present embodiment. FIG. 7 is a flowchart illustrating an example of a flow of a defect type determination process performed by the defect candidate area detection unit according to the present embodiment. FIG. 8 is a flowchart illustrating an example of a procedure flow of the feature value extraction process performed by the feature value extraction unit and the intensity value calculation process performed by the intensity calculation unit of the present embodiment.

  Hereinafter, embodiments of an inspection apparatus, an inspection method, and a program according to the present invention will be described in detail with reference to the accompanying drawings.

  FIG. 1 is a schematic diagram illustrating an example of a printed matter inspection system 1 according to the present embodiment. As shown in FIG. 1, the printed matter inspection system 1 includes a printing apparatus 100, a printed matter inspection apparatus 200 (an example of an inspection apparatus), and a stacker 300.

  The printing apparatus 100 includes an operation panel 101, photosensitive drums 103Y, 103M, 103C, and 103K, a transfer belt 105, a secondary transfer roller 107, a paper feeding unit 109, a conveyance roller pair 111, and a fixing roller 113. , And an inversion path 115.

  The operation panel 101 is an operation display unit that performs various operation inputs to the printing apparatus 100 and displays various screens.

  Each of the photosensitive drums 103Y, 103M, 103C, and 103K forms a toner image by performing an image forming process (charging process, exposure process, developing process, transfer process, and cleaning process), and the formed toner image. Is transferred to the transfer belt 105. In this embodiment, a yellow toner image is formed on the photoreceptor drum 103Y, a magenta toner image is formed on the photoreceptor drum 103M, a cyan toner image is formed on the photoreceptor drum 103C, and the photoreceptor drum 103K is formed. Although a black toner image is formed, the present invention is not limited to this.

  The transfer belt 105 conveys the toner image (full color toner image) transferred from the photosensitive drums 103Y, 103M, 103C, and 103K to the secondary transfer position of the secondary transfer roller 107. In this embodiment, a yellow toner image is first transferred to the transfer belt 105, and then a magenta toner image, a cyan toner image, and a black toner image are sequentially superimposed and transferred. Is not to be done.

  The paper feeding unit 109 accommodates a plurality of recording media in an overlapping manner, and feeds the recording media. Examples of the recording medium include, but are not limited to, recording paper, and any medium that can record an image, such as an OHP (Overhead Projector) sheet, may be used.

  The transport roller pair 111 transports the recording medium fed by the paper feed unit 109 in the direction of arrow s on the transport path a.

  The secondary transfer roller 107 collectively transfers the full-color toner image conveyed by the transfer belt 105 onto the recording medium conveyed by the conveyance roller pair 111 at the secondary transfer position.

  The fixing roller 113 fixes the full-color toner image on the recording medium by heating and pressing the recording medium on which the full-color toner image is transferred.

  In the case of single-sided printing, the printing apparatus 100 discharges a printed material, which is a recording medium on which a full-color toner image is fixed, to the printed material inspection apparatus 200. On the other hand, in the case of duplex printing, the printing apparatus 100 sends a recording medium on which a full-color toner image is fixed to the reverse path 115.

  The reverse path 115 reverses the front and back surfaces of the recording medium by switching back the sent recording medium and conveys it in the direction of the arrow t. The recording medium conveyed by the reversing path 115 is re-conveyed by the conveying roller pair 111, the full-color toner image is transferred to the surface opposite to the previous one by the secondary transfer roller 107, fixed by the fixing roller 113, and printed. The paper is discharged to the printed product inspection apparatus 200.

  The printed matter inspection apparatus 200 includes a reading unit 201 and an operation panel 203.

  The operation panel 203 is an operation display unit that inputs various operations to the printed material inspection apparatus 200 and displays various screens. Note that the operation panel 203 may be omitted. In this case, the operation panel 101 may also serve as the operation panel 203, or an externally connected PC (Personal Computer) may serve as the operation panel 203.

  The reading unit 201 electrically reads one surface of the printed matter discharged from the printing apparatus 100. The reading unit 201 can be realized by, for example, a line scanner. The printed matter inspection apparatus 200 may further include a reading unit that electrically reads the other surface of the printed matter. In this case, the reading unit that electrically reads the other surface of the printed material may have the same configuration as the reading unit 201. Then, the printed matter inspection apparatus 200 discharges the printed matter that has been read to the stacker 300.

  The stacker 300 includes a tray 311. The stacker 300 stacks the printed materials discharged by the printed material inspection apparatus 200 on the tray 311.

  FIG. 2 is a block diagram illustrating an example of a hardware configuration of the printing apparatus 100 according to the present embodiment. As shown in FIG. 2, the printing apparatus 100 has a configuration in which a controller 810 and an engine unit (Engine) 860 are connected by a PCI bus. The controller 810 is a controller that controls the entire control of the printing apparatus 100, drawing, communication, and input from the operation display unit 820. The engine unit 860 is an engine that can be connected to the PCI bus, and is, for example, a print engine such as a plotter. The engine unit 860 includes an image processing part such as error diffusion and gamma conversion in addition to the engine part.

  The controller 810 includes a CPU (Central Processing Unit) 811, a North Bridge (NB) 813, a system memory (MEM-P) 812, a South Bridge (SB) 814, a local memory (MEM-C) 817, and an ASIC. (Application Specific Integrated Circuit) 816 and a hard disk drive (HDD) 818, and the North Bridge (NB) 813 and the ASIC 816 are connected by an AGP (Accelerated Graphics Port) bus 815. The MEM-P 812 further includes a ROM 812a and a RAM 812b.

  The CPU 811 performs overall control of the printing apparatus 100, has a chip set including the NB 813, the MEM-P 812, and the SB 814, and is connected to other devices via this chip set.

  The NB 813 is a bridge for connecting the CPU 811 to the MEM-P 812, SB 814, and the AGP bus 815, and includes a memory controller that controls reading and writing of the MEM-P 812, a PCI master, and an AGP target.

  The MEM-P 812 is a system memory used as a memory for storing programs and data, a memory for developing programs and data, a memory for drawing printers, and the like, and includes a ROM 812a and a RAM 812b. The ROM 812a is a read-only memory used as a program / data storage memory, and the RAM 812b is a writable / readable memory used as a program / data development memory, a printer drawing memory, or the like.

  The SB 814 is a bridge for connecting the NB 813 to a PCI device and peripheral devices. The SB 814 is connected to the NB 813 via a PCI bus, and a network interface (I / F) unit and the like are also connected to the PCI bus.

  The ASIC 816 is an IC (Integrated Circuit) for image processing applications having hardware elements for image processing, and has a role of a bridge for connecting the AGP bus 815, the PCI bus, the HDD 818, and the MEM-C 817. The ASIC 816 includes a PCI target and an AGP master, an arbiter (ARB) that forms the core of the ASIC 816, a memory controller that controls the MEM-C 817, and a plurality of DMACs (Direct Memory) that rotate image data using hardware logic. Access Controller) and a PCI unit that performs data transfer between the engine unit 860 via the PCI bus. The ASIC 816 is connected to a USB 840 and an IEEE 1394 (the Institute of Electrical and Electronics Engineers 1394) interface (I / F) 850 via a PCI bus. The operation display unit 820 is directly connected to the ASIC 816.

  The MEM-C 817 is a local memory used as a copy image buffer and a code buffer, and the HDD 818 is a storage for storing image data, programs, font data, and forms.

  The AGP bus 815 is a bus interface for a graphics accelerator card proposed for speeding up graphics processing. The AGP bus 815 speeds up the graphics accelerator card by directly accessing the MEM-P812 with high throughput. It is.

  FIG. 3 is a block diagram illustrating an example of a hardware configuration of the printed matter inspection apparatus 200 according to the present embodiment. As shown in FIG. 3, the printed matter inspection apparatus 200 has a configuration in which a controller 910 and an engine unit (Engine) 960 are connected by a PCI bus. The controller 910 is a controller that controls the overall control, drawing, communication, and input from the operation display unit 920 of the printed matter inspection apparatus 200. The engine unit 960 is an engine that can be connected to the PCI bus, and is, for example, a scanner engine such as a scanner. The engine unit 960 includes an image processing part such as error diffusion and gamma conversion in addition to the engine part.

  The controller 910 includes a CPU 911, a north bridge (NB) 913, a system memory (MEM-P) 912, a south bridge (SB) 914, a local memory (MEM-C) 917, an ASIC 916, and a hard disk drive (HDD). 918, and the north bridge (NB) 913 and the ASIC 916 are connected by the AGP bus 915. The MEM-P 912 further includes a ROM 912a and a RAM 912b.

  The CPU 911 performs overall control of the printed matter inspection apparatus 200, has a chip set including the NB 913, the MEM-P 912, and the SB 914, and is connected to other devices via the chip set.

  The NB 913 is a bridge for connecting the CPU 911 to the MEM-P 912, SB 914, and the AGP bus 915, and includes a memory controller that controls reading and writing to the MEM-P 912, a PCI master, and an AGP target.

  The MEM-P 912 is a system memory used as a memory for storing programs and data, a memory for developing programs and data, a memory for drawing printers, and the like, and includes a ROM 912a and a RAM 912b. The ROM 912a is a read-only memory used as a memory for storing programs and data, and the RAM 912b is a writable and readable memory used as a program / data development memory, a printer drawing memory, and the like.

  The SB 914 is a bridge for connecting the NB 913 to a PCI device and a peripheral device. The SB 914 is connected to the NB 913 via a PCI bus, and a network interface (I / F) unit and the like are also connected to the PCI bus.

  The ASIC 916 is an IC for image processing that includes hardware elements for image processing, and has a role of a bridge that connects the AGP bus 915, the PCI bus, the HDD 918, and the MEM-C 917. The ASIC 916 includes a PCI target and an AGP master, an arbiter (ARB) that forms the core of the ASIC 916, a memory controller that controls the MEM-C 917, a plurality of DMACs that rotate image data using hardware logic, and the like. And a PCI unit that performs data transfer with the unit 960 via the PCI bus. The ASIC 916 is connected to a USB 940 and an IEEE 1394 interface (I / F) 950 via a PCI bus. The operation display unit 920 is directly connected to the ASIC 916.

  The MEM-C 917 is a local memory used as a copy image buffer and a code buffer, and the HDD 918 is a storage for storing image data, programs, font data, and forms.

  The AGP bus 915 is a bus interface for a graphics accelerator card proposed for speeding up graphics processing. The AGP bus 915 speeds up the graphics accelerator card by directly accessing the MEM-P 912 with high throughput. It is.

  FIG. 4 is a block diagram illustrating an example of functional configurations of the printing apparatus 100 and the printed matter inspection apparatus 200 according to the present embodiment. As illustrated in FIG. 4, the printing apparatus 100 includes a RIP (Raster Image Processor) unit 121, a print control unit 123, and a printing unit 125. The printed matter inspection apparatus 200 includes a reading unit 251, a read image acquisition unit 253, a reference image acquisition unit 255, a difference image generation unit 257, a storage unit 259, a defect candidate area detection unit 261, and a feature amount extraction unit 263. And an intensity calculation unit 265 and an inspection unit 267.

  In the present embodiment, the case where the printing apparatus 100 includes the RIP unit 121 will be described as an example. However, the present invention is not limited to this, and an apparatus different from the printing apparatus 100 such as DFE (Digital Front End) may include the RIP unit 121. You may make it prepare.

  In this embodiment, it is assumed that the printing apparatus 100 and the printed matter inspection apparatus 200 are connected by a local interface such as USB (Universal Serial Bus) or PCIe (Peripheral Component Interconnect Express). The connection form between the apparatus 100 and the printed matter inspection apparatus 200 is not limited to this.

  The RIP unit 121 and the print control unit 123 can be realized by the CPU 811 and the system memory 812, for example. The printing unit 125 is realized by, for example, the photosensitive drums 103Y, 103M, 103C, and 103K, the transfer belt 105, the secondary transfer roller 107, and the fixing roller 113, but is not limited thereto. As described above, in this embodiment, an image is printed by an electrophotographic method, but the present invention is not limited to this, and an image may be printed by an inkjet method.

  The reading unit 251 includes the reading unit 201 and can be realized by the engine unit 960, for example. The read image acquisition unit 253, the reference image acquisition unit 255, the difference image generation unit 257, the defect candidate area detection unit 261, the feature amount extraction unit 263, the intensity calculation unit 265, and the inspection unit 267 include, for example, a CPU 911 and a system memory 912. May be realized by ASIC 916 or the like, or may be realized by using these together. Note that the difference image generation unit 257, the defect candidate region detection unit 261, the feature amount extraction unit 263, the intensity calculation unit 265, and the inspection unit 267 may be realized by a GPU (Graphics Processing Unit) instead of the CPU 911. The storage unit 259 can be realized by the HDD 918, for example.

  The RIP unit 121 receives print data from an external device such as a host device, and generates an original image that is a print generation source (a basis for printing) from the received print data. Specifically, the RIP unit 121 performs RIP processing on the print data and generates an RIP image (bitmap image) as an original image.

  In the present embodiment, the print data includes data described in a page description language (PDL) such as PostScript (registered trademark), image data in a TIFF (Tagged Image File Format) format, and the like. However, the present invention is not limited to this. In the present embodiment, the original image is CMYK RIP image data. However, the present invention is not limited to this.

  The print control unit 123 transmits the original image generated by the RIP unit 121 to the printed matter inspection apparatus 200 and also to the printing unit 125.

  Further, the print control unit 123 uses the inspection result information transmitted (feedback) from the printed material inspection apparatus 200 to instruct the printing unit 125 to perform replacement printing or to perform quality inspection on the stacker 300, for example. For example, the paper discharge destination of the printed material that does not pass or the marking on the printed material that does not pass the quality inspection are performed.

  The printing unit 125 executes a printing process such as an image forming process, and generates a printed material by printing a print image based on the original image on a recording sheet.

  The reading unit 251 reads the printed matter generated by the printing unit 125 and generates a read image. Specifically, the reading unit 251 generates a read image by electronically reading the print image from a printed material on which a print image based on the original image is printed. In the present embodiment, the read image is RGB image data, and each pixel of the R, G, and B image data is 8-bit 200 dpi. However, the present invention is not limited to this.

  The read image acquisition unit 253 acquires a read image obtained by reading a printed material. Specifically, the read image acquisition unit 253 acquires the read image generated by the reading unit 251.

  The reference image acquisition unit 255 acquires a reference image serving as an inspection standard for printed matter. In the present embodiment, the reference image acquisition unit 255 acquires an original image from the printing apparatus 100 and generates a reference image (master image) based on the acquired original image, thereby acquiring the reference image. Specifically, the reference image acquisition unit 255 acquires RIP images of C, M, Y, and K from the printing apparatus 100 (print control unit 123), and acquires the acquired RIPs of C, M, Y, and K, respectively. The image is subjected to various image processing such as multi-value conversion processing, smoothing processing, resolution conversion processing, and color conversion processing to generate a reference image. In the present embodiment, the reference image is RGB image data, and each pixel of the R, G, and B image data is 8-bit 200 dpi. However, the present invention is not limited to this. Note that the reference image acquisition unit 255 may acquire the original image transmitted from the printing apparatus 100 as a reference image.

  The difference image generation unit 257 generates a difference image indicating a difference between the read image acquired by the read image acquisition unit 253 and the reference image acquired by the reference image acquisition unit 255. Specifically, the difference image generation unit 257 performs alignment between the acquired read image and the reference image, compares the read image after the alignment with the reference image on a pixel basis, and compares the pixel values of 8-bit RGB colors. The difference value is calculated for each pixel, and a difference image composed of the difference value of the pixel value for each pixel is generated.

  The defect candidate area detection unit 261 detects a defect candidate area including a defect candidate part that is a defect candidate on the difference image generated by the difference image generation unit 257. Specifically, the defect candidate area detection unit 261 detects the defect candidate area on the difference image using the threshold information for defect candidate detection stored in the storage unit 259.

  FIG. 5 is a diagram illustrating an example of threshold information for defect candidate detection according to the present embodiment. The density difference threshold value of the defect candidate is a threshold value for each pixel of the difference image, and a pixel exceeding this threshold value is determined as an abnormal pixel. When there are a plurality of abnormal pixels within a certain range, they are treated as one abnormal pixel group. The area threshold of the defect candidate is a threshold for the number of abnormal pixels included in one abnormal pixel group. When the number of abnormal pixels included in one abnormal pixel group exceeds this threshold, the abnormal pixel group Is determined as a defect candidate part to be a defect candidate.

  As described above, the defect candidate area detection unit 261 performs the determination using the defect candidate density difference threshold on the difference image, and determines the defect candidate for the determination result using the defect candidate density difference threshold. By performing the determination using the area threshold, a defect candidate part is detected, and a circumscribed rectangle circumscribing the defect candidate part is detected as a defect candidate region.

  As a result, the coordinates of the defect candidate area (for example, the upper left x and y coordinates and the lower right x and y coordinates of the defect candidate area) and the number of abnormal pixels included in the defect candidate part are obtained.

  The defect candidate area detection unit 261 performs image processing for conspicuous defect candidate sites such as smoothing processing or sharpening processing, or VTF (Visual Transfer Function) for the difference image in order to increase detection accuracy of the defect candidate sites. ) After performing a filtering process using a function that functions as a filter that obtains a spatial frequency corresponding to the human visual characteristic that strongly reacts to a specific frequency component, such as a function, a defect candidate site may be detected. . In particular, if such a filtering process is performed, the calculation accuracy of an intensity value representing the visibility of a defect candidate site described later can be increased.

  And the defect candidate area | region detection part 261 determines the defect classification of the detected defect candidate site | part. In the present embodiment, the defect candidate area detection unit 261 determines whether the detected defect type of the defect candidate part is a point defect or a line defect, but is not limited thereto. .

  In addition, whether the defect type of the detected defect candidate part is a point defect or a linear defect is determined by, for example, the aspect ratio (vertical and horizontal length) of the defect candidate area (circumscribed rectangle) of the detected defect candidate part. Ratio). In the present embodiment, the defect candidate region detection unit 261 detects the defect in the detected defect candidate region if the aspect ratio of the defect candidate region of the detected defect candidate region is equal to or greater than a dotted / linear identification threshold (for example, 5). If it is determined that the type is a linear defect and the aspect ratio is less than the dotted / linear identification threshold, it is determined that the defect type of the detected defect candidate site is a point defect. It is not limited to. Note that the dotted / linear identification threshold is stored in the storage unit 259. The point / line identification threshold may be set as separate thresholds for identifying a point defect and a line identification threshold for identifying the line defect.

  Thereby, the defect type of a defect candidate site | part is obtained.

  The feature amount extraction unit 263 extracts at least one of the color feature amount and the shape feature amount of the defect candidate part detected by the defect candidate region detection unit 261. Specifically, the feature amount extraction unit 263 extracts a color feature amount and a shape feature amount according to the defect type of the defect candidate part detected by the defect candidate region detection unit 261.

  For example, when the defect type of the defect candidate part detected by the defect candidate area detection unit 261 is a point defect, the feature amount extraction unit 263 extracts the maximum value of the color difference of the defect candidate area as a color feature amount, As the shape feature amount, the number of pixels of the defect candidate portion (specifically, the number of abnormal pixels) is extracted.

  In addition, for example, when the defect type of the defect candidate part detected by the defect candidate region detection unit 261 is a linear defect, the feature amount extraction unit 263 extracts an average value of color differences of the defect candidate regions as a color feature amount. Then, the number of pixels having the width of the defect candidate part is extracted as the shape feature amount.

In the present embodiment, the color difference ΔE ₀₀ of CIEDE 2000 is described as an example of the color difference of the color feature amount, but the present invention is not limited to this. The color difference ΔE ₀₀ of CIEDE 2000 is a value that quantitatively represents a perceptual difference in color.

The feature amount extraction unit 263 converts the read image and the inspection image from the RGB color system to the CIE Lab color system in order to obtain the color difference ΔE ₀₀ of the defect candidate part. Since the conversion formula from the RGB color system to the CIE Lab color system is a known technique, a detailed description thereof is omitted. The CIE Lab color system is a color system in which a color difference that quantitatively represents a perceptual difference in color is intended to correspond to a distance in the color space, and represents lightness L * and hue and saturation. It is a color system in which chromaticity is represented by a * and b *.

Next, the feature amount extraction unit 263 obtains a color difference ΔE ₀₀ for each pixel constituting the defect candidate area using the Lab value of the defect candidate area on the read image and the Lab value of the defect candidate area on the reference image. . Since the defect candidate area coordinates are obtained by the defect candidate area detection unit 261, the location of the defect candidate area can be identified from the read image and the reference image.

Since the calculation formula for the color difference ΔE ₀₀ is a known technique, a detailed description thereof will be omitted. It should be noted that the human eye has a range called a color identification area in which the color difference cannot be identified even though the colors are different. The calculation formula for the color difference ΔE ₀₀ is that the color difference based on the calculation is the Lab three-dimensional color space. The calculation formula is defined so as to approximate the color discrimination range of the human eye. In addition, the CIE Lab color system was intended to express color differences that quantitatively represent perceptual differences in colors in a uniform distance space. There is a problem that the color difference expressed by the above and the evaluation by the human eye are different.

For this reason, although details will be described later, the color difference ΔE ₀₀ is used to calculate the intensity value indicating the visibility of the defect candidate site as in the present embodiment, so that the human eye due to the difference in the color and brightness of the reference image. Therefore, the intensity value can be obtained without requiring a weighting coefficient for each color.

Next, the feature amount extraction unit 263 obtains the maximum value and the average value of the color difference ΔE ₀₀ of the defect candidate area from the color difference ΔE ₀₀ calculated for each pixel constituting the defect candidate area, and the defect candidate part included in the defect candidate area If the defect type is a point defect, the maximum value of the color difference ΔE ₀₀ is extracted (adopted) as the color feature amount of the defect candidate part, and the defect type of the defect candidate part included in the defect candidate area is a linear defect. If there is, the average value of the color difference ΔE ₀₀ is extracted (adopted) as the color feature amount of the defect candidate part.

Note that when the printed matter to be inspected is uniform, the feature amount extraction unit 263 obtains the maximum value and the average value for each of the RGB read image and the reference image without obtaining the color difference ΔE ₀₀ for all the pixels in the defect candidate region. The maximum value and average value of the obtained read image and reference image may be converted into Lab color system Lab values to obtain the color difference ΔE ₀₀ .

  Further, as described above, if the defect type of the defect candidate region is a point defect, the feature amount extraction unit 263 extracts (adopts) the number of pixels of the defect candidate portion as the shape feature amount of the defect candidate portion. If the defect type of the candidate area is a linear defect, the number of pixels having the width of the defect candidate part is extracted (adopted) as the shape feature amount of the defect candidate part.

  In the present embodiment, the number of pixels in the width of the defect candidate region is not the number of pixels in the width of the defect candidate area itself, but the average value of the number of abnormal pixels in each short side direction. This is because, for example, the width of the defect candidate portion is appropriately evaluated even when the width of only a part of the linear defect is thick or the case of a linear defect that crosses the paper diagonally.

  The intensity calculation unit 265 calculates an intensity value representing the visibility of the defect candidate part based on at least one of the color feature value and the shape feature value extracted by the feature value extraction unit 263. In the present embodiment, the intensity calculation unit 265 calculates the intensity value of the defect candidate site using Equation (1).

Intensity value = log ₁₀ (ΔE ₀₀ ⁿ × S ^m ) (1)

Here, ΔE ₀₀ represents a color difference which is a color feature amount. S represents the number of pixels which is a shape feature amount. n is a coefficient for weighting the color feature amount, and represents an arbitrary real number. m is a coefficient for weighting the shape feature value, and represents an arbitrary real number.

Here, the point defect is often relatively small, and the boundary value of the defect may become dull due to the influence of optical or image processing. In order to reduce the influence, in this embodiment, the intensity value Is calculated, the maximum value of the color difference ΔE ₀₀ is adopted as the color component (color difference ΔE ₀₀ ). However, even in the case of using the average value of the color difference Delta] E _00, compared with the case of using the maximum value of the color difference Delta] E _00, since a large difference does not occur, it is also possible to use the average value of the color difference Delta] E ₀₀ .

  In the case of a point-like defect, the number of pixels represents the size of the defect, and the larger one is more conspicuous as a defect and higher in visibility. Therefore, in this embodiment, when calculating the intensity value, the shape component S is used as the shape component S. The number of pixels of the defect candidate part is adopted.

In addition, when a line defect has a point defect on a part of a defect area, the value of the point may increase specifically, and using the maximum value of the color difference ΔE ₀₀ will greatly affect the influence. Therefore, in this embodiment, when calculating the intensity value, the average value of the color difference ΔE ₀₀ is used as the color component (color difference ΔE ₀₀ ).

  In the case of a linear defect, the length does not significantly affect the visibility, but the width greatly affects the visibility. In this embodiment, when calculating the intensity value, the defect candidate site is used as the shape component S in this embodiment. The number of pixels of the width is adopted. However, since the length does not affect the visibility at all, the number of pixels corresponding to the length of the defect candidate part may be employed as the shape feature amount of the defect candidate part. For example, the number of pixels of the width of the defect candidate portion may be multiplied by a coefficient that decreases when the length is short and increases when the length is long.

  The inspection unit 267 inspects whether the defect candidate site is a defect based on the intensity value calculated by the intensity calculation unit 265. For example, the inspection unit 267 determines that the defect candidate part is a defect if the intensity value of the defect candidate part calculated by the intensity calculation unit 265 is greater than or equal to an intensity threshold (for example, 3), and less than the intensity threshold. The defect candidate part is determined not to be a defect. The intensity threshold value is stored in the storage unit 259. Further, for example, the inspection unit 267 may rank the defect degree of the defect candidate part into a plurality of levels (for example, 10 levels) based on the intensity value of the defect candidate part calculated by the intensity calculation unit 265.

  Then, the inspection unit 267 associates inspection results such as color feature amounts, shape feature amounts, intensity values, positions, and defect types of defect candidate portions determined to be defects, read images, reference images, difference images, and the like. The data is stored in the HDD 918 or transmitted (feedback) to the printing apparatus 100.

  FIG. 6 is a flowchart illustrating an example of a flow of processing performed by the printed matter inspection apparatus 200 according to the present embodiment.

  First, the read image acquisition unit 253 acquires the read image generated by the reading unit 251 (step S101).

  Subsequently, the reference image acquisition unit 255 acquires an original image from the printing apparatus 100 and generates a reference image (master image) based on the acquired original image, thereby acquiring the reference image (step S103).

  Subsequently, the difference image generation unit 257 generates a difference image indicating a difference between the read image acquired by the read image acquisition unit 253 and the reference image acquired by the reference image acquisition unit 255 (step S105).

  Subsequently, the defect candidate area detection unit 261 detects a defect candidate area including a defect candidate part that is a defect candidate on the difference image generated by the difference image generation part 257, and the defect of the detected defect candidate part is detected. The type is determined (step S107).

  Subsequently, the feature amount extraction unit 263 extracts a color feature amount and a shape feature amount according to the defect type of the defect candidate part detected by the defect candidate region detection unit 261 (step S109).

  Subsequently, the intensity calculating unit 265 calculates an intensity value representing the visibility of the defect candidate site based on the color feature value and the shape feature value extracted by the feature value extracting unit 263 (step S111).

  Subsequently, the inspection unit 267 inspects whether the defect candidate site is a defect based on the intensity value calculated by the intensity calculation unit 265 (step S113).

  FIG. 7 is a flowchart illustrating an example of a flow of a defect type determination process performed by the defect candidate area detection unit 261 of the present embodiment.

  First, the defect candidate area detection unit 261 determines whether or not the aspect ratio of the defect candidate area of the detected defect candidate part is equal to or greater than the dotted / linear identification threshold (step S201).

  If the aspect ratio of the defect candidate area is equal to or greater than the dotted / linear identification threshold (Yes in step S201), the defect type is a linear defect, so the defect candidate area detection unit 261 indicates the width of the defect candidate part. It is determined whether or not the thickness of the line is equal to or greater than the band / muscle identification threshold (step S203).

  When the thickness of the line, which is the width of the defect candidate part, is equal to or greater than the band / muscle identification threshold (Yes in step S203), the defect candidate area detection unit 261 sets the defect type of the defect candidate part in one form of a linear defect. It is determined as a certain band defect (step S205).

  On the other hand, when the thickness of the line, which is the width of the defect candidate part, is less than the band / muscle identification threshold (No in step S203), the defect candidate area detection unit 261 sets the defect type of the defect candidate part as one of the linear defects. It is determined that the defect is a muscle defect (step S207).

  In addition, when the aspect ratio of the defect candidate area is less than the dotted / linear identification threshold value (No in step S201), the defect type is a point defect, so that the defect candidate area detecting unit 261 determines the pixel of the defect candidate part. It is determined whether or not the number is greater than or equal to the plaque / spot identification threshold (step S209).

  When the number of pixels of the defect candidate part is equal to or greater than the spot / point identification threshold (Yes in step S209), the defect candidate area detection unit 261 determines that the defect type of the defect candidate part is a spot defect that is one aspect of the point defect. (Step S211).

  On the other hand, when the number of pixels of the defect candidate part is less than the spot / point identification threshold (No in step S209), the defect candidate area detection unit 261 sets the defect type of the defect candidate part as a point defect that is an aspect of a point defect. Is determined (step S213).

  FIG. 8 is a flowchart illustrating an example of a procedure flow of the feature value extraction process performed by the feature value extraction unit 263 and the intensity value calculation process performed by the intensity calculation unit 265 of the present embodiment.

  First, the feature amount extraction unit 263 converts the read image and the inspection image from the RGB color system to the CIE Lab color system (step S301).

Subsequently, the feature amount extraction unit 263 uses the Lab value of the defect candidate area on the read image and the Lab value of the defect candidate area on the reference image to calculate the color difference ΔE ₀₀ for each pixel constituting the defect candidate area. Obtain (step S303).

Subsequently, the feature amount extraction unit 263 calculates the maximum value and the average value of the color difference ΔE ₀₀ of the defect candidate region from the color difference ΔE ₀₀ calculated for each pixel constituting the defect candidate region (step S305).

Subsequently, if the defect type of the defect candidate part included in the defect candidate area is a point defect (Yes in step S307), the feature quantity extraction unit 263 uses the color difference as the color feature quantity (color difference component) of the defect candidate part. The maximum value of ΔE ₀₀ is set (step S309), and the number of pixels of the defect candidate part is set as the shape feature amount (shape component) of the defect candidate part (step S311).

On the other hand, if the defect type of the defect candidate portion included in the defect candidate region is a linear defect (No in step S307), the feature amount extraction unit 263 uses the color difference ΔE as the color feature amount (color difference component) of the defect candidate portion. An average value of ₀₀ is set (step S313), and the number of pixels in the width of the defect candidate portion is set as the shape feature amount (shape component) of the defect candidate portion (step S315).

  Subsequently, the intensity calculation unit 265 calculates the intensity value of the defect candidate part based on the color feature quantity (color difference component) and the shape feature quantity (shape component) set by the feature quantity extraction unit 263 (step S317). ).

  As described above, according to the present embodiment, whether or not a defect candidate part is a defect is inspected based on an intensity value calculated based on at least one of a color feature amount and a shape feature quantity of the defect candidate part. Therefore, it becomes possible to inspect various defects with high accuracy. In detail, since the color feature amount and the shape feature amount corresponding to the defect type of the defect candidate part are adopted, it becomes possible to inspect various defects with high accuracy.

(Modification)
In the above-described embodiment, the online type printed matter inspection system in which the printing apparatus 100 and the printed matter inspection apparatus 200 are integrated has been described as an example, but the offline state in which the printing apparatus 100 and the printed matter inspection apparatus 200 are separated. It may be a printed matter inspection system of a mold.

  In the above-described embodiment, the case where the printed matter inspection apparatus 200 is an apparatus dedicated to printed matter inspection is described as an example. .

  In the above embodiment, the case where the inspection target is a printed material has been described as an example. However, the inspection target is not limited to the printed material as long as the read image and the reference image of the inspection target can be acquired.

  In the above-described embodiment, an RGB image is assumed as a read image. For example, a filter that transmits light in a specific wavelength band and a monochrome camera are combined to change the light from each wavelength band to a target color system. It may be converted and inspected.

(program)
The program executed by the printed matter inspection apparatus according to the above-described embodiment and each modified example is a file in an installable or executable format, and is a CD-ROM, CD-R, memory card, DVD (Digital Versatile Disk), flexible disk. The program is stored in a computer-readable storage medium such as (FD).

  In addition, the program executed by the printed matter inspection apparatus 200 of the above-described embodiment and each modification may be provided by being stored on a computer connected to a network such as the Internet and downloaded via the network. Further, the program executed by the printed matter inspection apparatus 200 according to the embodiment and each modification may be provided or distributed via a network such as the Internet. Further, the program executed by the printed matter inspection apparatus 200 according to the above embodiment and each modification may be provided by being incorporated in advance in a ROM or the like.

  The program executed by the printed matter inspection apparatus 200 according to the embodiment and each modified example has a module configuration for realizing the above-described units on a computer. As actual hardware, for example, the CPU reads out a program from the ROM to the RAM and executes the program, so that each functional unit is realized on the computer.

  In addition, the said embodiment and modification are shown as an example and are not intending limiting the range of invention. The above-described novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. These embodiments are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

DESCRIPTION OF SYMBOLS 1 Printed product inspection system 100 Printing apparatus 101 Operation panel 103Y, 103M, 103C, 103K Photosensitive drum 105 Transfer belt 107 Secondary transfer roller 109 Paper feed part 111 Conveying roller pair 113 Fixing roller 115 Reverse path 121 RIP part 123 Print control part 125 Printing unit 200 Printed material inspection apparatus 201 Reading unit 203 Operation panel 251 Reading unit 253 Read image acquisition unit 255 Reference image acquisition unit 257 Difference image generation unit 259 Storage unit 261 Defect candidate area detection unit 263 Feature quantity extraction unit 265 Strength calculation unit 267 Inspection Part 300 Stacker 301 Tray

Japanese Patent Laying-Open No. 2015-179525

Claims (6)

A read image acquisition unit that acquires a read image obtained by reading a printed matter;
A reference image acquisition unit for acquiring a reference image serving as an inspection reference for the printed matter;
A difference image generation unit that generates a difference image indicating a difference between the read image and the reference image;
On the difference image, a defect candidate area detection unit that detects a defect candidate area including a defect candidate part that is a defect candidate; and
A feature amount extraction unit that extracts at least one of a color feature amount and a shape feature amount of the defect candidate portion;
Based on at least one of the color feature amount and the shape feature amount, an intensity value calculation unit that calculates an intensity value representing the visibility of the defect candidate site;
Based on the intensity value, an inspection unit for inspecting whether the defect candidate site is a defect,
An inspection apparatus comprising: The defect candidate area detection unit determines a defect type of the detected defect candidate part,
The inspection apparatus according to claim 1, wherein the feature amount extraction unit extracts a color feature amount and a shape feature amount according to the defect type of the defect candidate part. The defect candidate area detection unit determines whether or not the defect type of the detected defect candidate part is a point defect,
When the defect type of the defect candidate part is the point defect, the feature amount extraction unit extracts the maximum value of the color difference of the defect candidate region as the color feature amount, and as the shape feature amount, The inspection apparatus according to claim 2, wherein the number of pixels of the defect candidate part is extracted. The defect candidate area detection unit determines whether the defect type of the detected defect candidate part is a linear defect,
When the defect type of the defect candidate part is the linear defect, the feature amount extraction unit extracts an average value of color differences of the defect candidate regions as the color feature amount, and as the shape feature amount, The inspection apparatus according to claim 2, wherein the number of pixels having a width of a defect candidate part is extracted. A read image acquisition step of acquiring a read image obtained by reading a printed matter;
A reference image acquisition step of acquiring a reference image to be an inspection reference of the printed matter;
A difference image generation step of generating a difference image indicating a difference between the read image and the reference image;
On the difference image, a defect candidate region detection step for detecting a defect candidate region including a defect candidate portion that is a defect candidate; and
A feature amount extracting step of extracting at least one of a color feature amount and a shape feature amount of the defect candidate part;
An intensity value calculating step for calculating an intensity value representing the visibility of the defect candidate part based on at least one of the color feature quantity and the shape feature quantity;
An inspection step for inspecting whether the defect candidate site is a defect based on the intensity value;
Including inspection methods. A read image acquisition step of acquiring a read image obtained by reading a printed matter;
A reference image acquisition step of acquiring a reference image to be an inspection reference of the printed matter;
A difference image generation step of generating a difference image indicating a difference between the read image and the reference image;
On the difference image, a defect candidate region detection step for detecting a defect candidate region including a defect candidate portion that is a defect candidate; and
A feature amount extracting step of extracting at least one of a color feature amount and a shape feature amount of the defect candidate part;
An intensity value calculating step for calculating an intensity value representing the visibility of the defect candidate part based on at least one of the color feature quantity and the shape feature quantity;
An inspection step for inspecting whether the defect candidate site is a defect based on the intensity value;
A program that causes a computer to execute.

Images