JP2016035418A - Image processing device, output object inspection method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing device, an output object inspection method, and a program for inspecting an output object from an output device.SOLUTION: An image processing device 150 includes: inspection image acquisition means 174 for acquiring an inspection image generated by reading an output object from an output device 110 as an image with a prescribed resolution; reference image generation means 170 for generating a reference image having approximately the same resolution as the inspection image on the basis of the data that is the source of the output object; dividing means 210 and 212 for dividing each of the reference and inspection images into area units; measurement means 214 for measuring an edge characteristic for each unit area of the reference image; reduction processing means 218 and 220 for reducing each of the area-unit image of the reference image whose edge characteristic is below normal and the corresponding area-unit image of the inspection image; and defect detection means 222 for comparing, with respect to the area unit whose edge characteristic is below normal, the reduced images of the reference and inspection images and detecting a defect.SELECTED DRAWING: Figure 7

Description

  The present invention relates to an image processing apparatus, an image inspection method, and a program, and more particularly to an image processing apparatus, an output object inspection method, and a program for inspecting an output from an output device.

  Conventionally, various inspections are required for printed output in the field of production and printing. In response to such demands, there is provided a printed matter inspection apparatus that reads an output product output from an image forming apparatus with a line sensor of a camera / scanner and inspects whether the printing is normally performed based on the read result. Yes.

  In the printed matter inspection apparatus, particularly when performing print quality inspection, an inspection image obtained by image reading, a RIP (Raster Image Processer) of original image data prepared by a user, a master image assuming printing and reading, and A comparison is made. Then, after accurately aligning the inspection image and the master image, the presence or absence of a defect is determined based on the degree of difference between the inspection image and the master image. For example, a difference is obtained for each pixel, and a defect is determined when the difference value exceeds a predetermined determination threshold.

  A technique is also known in which both the inspection image and the master image are divided into blocks, the positions of each block are corrected and combined, and the pixels are compared and the difference is detected with high accuracy (for example, a special feature). No. 2013-186562). Due to the settings and characteristics of the printer output, the master image and the print result are not output in exactly the same size in the first place, or because of the expansion and contraction due to the humidity of the printed paper and the meandering of the transport device, This corresponds to the change of the angle.

  In print inspection, even when printing is performed at a relatively high resolution of 1200 dpi, an inspection standard based on human sensitivity evaluation may not require such a high inspection resolution. In order to improve the inspection processing speed, it is desirable to be able to inspect with the minimum low resolution and to satisfy the inspection standard. For this reason, in reality, it is desired that a resolution for satisfying the standard is obtained experimentally, and a printed matter inspection apparatus with a relatively low resolution of, for example, 200 dpi can be configured.

  However, the inspection resolution obtained experimentally in this way is almost always determined on an average and a large scale, and there is a rare case where a shortage of inspection occurs in a special print original. For example, artistic line drawings, which are not included in normal natural images and text images, and registration marks for positioning necessary for the bookbinding of printed materials, etc., tend to be insufficiently inspected.

  In Japanese Patent No. 5308735 (Patent Document 2), by converting the resolution of the target image in accordance with the resolution of the image to be inspected, it is possible to efficiently match the size of a halftone dot or convert from CMYK to RGB. A technique for converting to a size is disclosed. Japanese Patent No. 4834387 (Patent Document 3) is capable of reducing the resolution of the entire image by a technique of inspecting a halftone-processed image by changing inspection conditions according to the type of the recording medium and the “simple mode”. Is mentioned.

  Furthermore, Japanese Patent Laid-Open No. 2013-046092 (Patent Document 4) compares image data subjected to image processing for printing with read data obtained by an image reading apparatus, and an image printed on paper is defective. In the control unit for determining whether or not the image data is attributed, a configuration for switching the resolution to be converted for each attribute of the image based on the attribute data of the image data is disclosed.

  If the inspection resolution of the entire image is increased so that there is no lack of inspection for images that require high-resolution inspections that are infrequent, the burden of inspection processing will increase, reducing inspection processing time and reducing costs. This is a difficult point. Similarly, as disclosed in Patent Document 2 and Patent Document 3, the method of applying a resolution that can detect all defects in one image increases inspection processing time and apparatus cost.

  In addition, since the prior art of Patent Document 4 gives semantic attributes to important character strings and numbers included in image data and is based on these attributes, it is not necessarily matched with the image characteristics of an actual image. It was not something that could be switched. Further, in the prior art of Patent Document 4, since an attribute is assigned to each pixel, it is not sufficient in that the logic for comparison and inspection becomes complicated if the resolution is changed for each region to which the attribute is attached. .

  The present invention has been made in view of the insufficiency in the prior art described above, and the present invention provides an image inspection process in which an output product from an output device is scanned and inspected at a rare high resolution. Provided are an image processing apparatus, an output object inspection method, and a program capable of identifying an image portion that needs to be inspected and improving an overall average processing speed while avoiding a lack of inspection in the image portion. For the purpose.

  In order to solve the above-described problems, the present invention provides an image processing apparatus having the following characteristics. The image processing apparatus includes: an inspection image acquisition unit that acquires an inspection image generated by reading an output object from the output device at a predetermined resolution; and the inspection image based on data that is a source of the output object. A reference image generating means for generating a reference image having substantially the same resolution as the above, a dividing means for dividing the reference image and the inspection image into area units, and measurement for measuring edge characteristics for each area unit of the reference image Means, a reduction processing means for reducing an image of a region unit of a reference image whose edge characteristics are less than a specified value, and a corresponding region unit image of an inspection image, and the reference image with respect to a region unit whose edge properties are less than a specified value And defect detection means for detecting defects by comparing the reduced images of the inspection image.

  With the above configuration, in an image inspection process in which an image output from an output device is scanned and inspected, an image portion that rarely needs to be inspected at a high resolution is identified, and an insufficient inspection in the image portion is avoided. However, the overall average processing speed can be improved.

The mechanism block diagram of the printing system containing the printed matter inspection apparatus by this embodiment. 1 is a hardware block diagram of a printing system including a printed material inspection apparatus according to an embodiment. FIG. 5 is a block diagram illustrating the operation of the printed matter inspection apparatus according to the present embodiment (in a paper registration mode). The block diagram which shows operation | movement of the printed matter inspection apparatus by this embodiment (at the time of normal mode). The figure which shows the data structure of (A) job management information and (B) paper correction information. The detailed block diagram of the master image generation part by this embodiment. The detailed block diagram of the image defect determination part by this embodiment. The figure explaining the aspect which divides | segments the whole image into a some block in this embodiment. 6 is a flowchart illustrating image inspection processing executed by the printed matter inspection apparatus according to the embodiment.

  Hereinafter, although this embodiment is described, the embodiment is not limited to the embodiment described below. In the embodiment described below, an image processing apparatus that reads and inspects an output product from an output device is connected to a subsequent stage of the printing apparatus and reads and inspects the printed matter output from the printing apparatus. A printed material inspection apparatus will be described as an example.

  FIG. 1 is a diagram showing a mechanism configuration of a printing system including a printed matter inspection apparatus according to the present embodiment. A printing system 10 shown in FIG. 1 includes a printing apparatus 20, a printed matter inspection apparatus 50 according to the present embodiment, and a post-processing apparatus 90.

  In response to receiving a print job including a print image from the outside, the printing apparatus 20 issues an instruction to execute the print job stored in the printing apparatus 20 via the operation panel 40 included in the printing apparatus 20. In response to the reception, an image is formed on a transfer member such as paper according to the contents of the print job, and is printed out.

  In FIG. 1, the sheet conveyance path is indicated by a dotted line. First, the paper is taken out by the paper feed unit 32 and fed to the transfer site by the paper feed roller 34. The printing apparatus 20 includes a photosensitive drum 22 for each color. In the embodiment shown in FIG. 1, a tandem system in which four color photosensitive drums 22K, 22C, 22M, and 22Y of black (K), cyan (C), magenta (M), and yellow (Y) are arranged in parallel. It is configured.

  An electrostatic latent image is formed on the surface of each photoconductive drum 22 by controlling on / off of the light beam in accordance with writing signals of each color of K, C, M, and Y generated based on the print image data. . The formed electrostatic latent image is conveyed toward the developing device as the photosensitive drum 22 rotates and is developed with toner. A toner image corresponding to the writing signal is formed on the photosensitive drum 22 and carried.

  On the other hand, an intermediate transfer belt 30 stretched around conveying rollers 24 and 26 is disposed below the photosensitive drums 22K, 22C, 22M, and 22Y. The toner image is conveyed toward the intermediate transfer belt 30 as the photosensitive drum 22 rotates, and is transferred onto the intermediate transfer belt 30 at a position where the photosensitive drum 22 and the intermediate transfer belt 30 are in contact with each other by a primary transfer roller. . The toner images of the respective colors are superimposed at the same position, and a multicolor toner image is formed on the intermediate transfer belt 30. The multicolor toner image transferred onto the intermediate transfer belt 30 is conveyed toward the roller 28 of the secondary transfer portion, and transferred onto the paper fed to the position of the transfer site by the transfer roller.

  The sheet on which the multicolor toner image is transferred is heated and fixed by the fixing roller 36. In the case of single-sided printing, the sheet is discharged as it is to the printed matter inspection apparatus 50. On the other hand, in the case of duplex printing, after being reversed by the reversing path 38, the toner image is further transferred and fixed on the opposite surface of the sheet, and discharged to the printed matter inspection apparatus 50.

  The printed matter inspection device 50 operates in cooperation with the printing device 20 and inspects the printed matter output by the printing device 20. A printed matter discharged from the printing device 20 is fed to the printed matter inspection device 50. The printed matter is image-read on both sides by the image reading devices 52 and 54, inspected, and then discharged to the post-processing device 90.

  As shown in FIG. 1, the printed matter inspection apparatus 50 may include an operation panel 56, but is not particularly limited, and the operation panel 40 of the printing apparatus 20 is also used without providing the operation panel 56. Also good. In addition to providing a dedicated printed matter inspection apparatus at the subsequent stage of the printing apparatus 20, a general-purpose computer receives read image data from an image reading apparatus provided in a paper discharge unit of the printing apparatus via a network such as a LAN (Local Area Network). It is good also as an aspect which transmits to and performs the test | inspection by image processing on a general purpose computer.

  The post-processing device 90 is, for example, a stacker, and stacks the inspected printed matter discharged from the printed matter inspection device 50 on the tray 92.

  FIG. 2 is a diagram illustrating hardware blocks of the printing system including the printed matter inspection apparatus according to the present embodiment.

  The printing apparatus 110 includes a system control unit 112, a user interface (hereinafter, the interface may be simply referred to as I / F) unit 118, a DFE (digital front end) I / F unit 120, A network I / F unit 122, an external I / F unit 124, an HDD (Hard Disk Drive) 126, an image processing control unit 128, a printing control unit 130, and a mechanism control unit 132 are included.

  The system control unit 112 includes a CPU 114 and a memory 116 inside, and functions as a control unit that controls the entire printing apparatus 110. The user I / F unit 118 is an interface for connecting the system control unit 112 and the operation panel 40. The network I / F unit 122 is an interface for connecting the system control unit 112 to a network such as a LAN or Ethernet (registered trademark). The external I / F unit 124 is an interface for connecting the system control unit 112 and an external device such as the printed matter inspection device 50. The HDD 126 is a storage device that stores a control program, a print job, and image data.

  The mechanism control unit 132 is a control unit that controls printing operations in the printing apparatus 110 such as paper conveyance and transfer processes in the printing apparatus 110. An external image generation controller 12 is connected to the DFEI / F unit 120. The DFEI / F unit 120 is an interface for receiving a transfer of an image subjected to RIP processing (hereinafter sometimes referred to as a RIP image) from an external image generation controller 12 connected thereto. The image processing control unit 128 controls print image processing in which the mechanism control unit 132 prepares a print image to be transferred to a sheet. The image processing control unit 128 also transfers the print image prepared by the image processing control unit 128 to the printed matter inspection apparatus 50 connected via the external I / F unit 124. The printing control unit 130 controls image formation on a transfer member such as paper.

  In contrast, the printed matter inspection apparatus 150 includes a system control unit 152, an external I / F unit 158, a user I / F unit 160, an HDD 162, a network I / F unit 164, a mechanism control unit 168, A master image generation unit 170, an image defect determination unit 172, and an image reading unit 174 are included.

  The system control unit 152 includes a CPU 154 and a memory 156 therein, and functions as a control unit that controls the entire printed matter inspection apparatus 150. The external I / F unit 158 is an interface for connecting the system control unit 152 and external devices such as the printing apparatus 110 and the post-processing apparatus 90. The user I / F unit 160 is an interface for connecting the system control unit 152 and the operation panel 56. The HDD 162 is a storage device that stores control programs and image data. The network I / F unit 164 is an interface that connects the system control unit 152 to a network such as a LAN or Ethernet (registered trademark).

  The mechanism control unit 168 functions as a control unit for a conveyance operation of the printed matter inspection apparatus 150 such as conveyance of a printed matter that is an inspection target. The master image generation unit 170 generates a master image serving as a reference image for inspection based on a print image sent from the printing apparatus 110 via an interface such as the external I / F unit 158. Reference image generating means is configured. The image reading unit 174 constitutes an inspection image acquisition unit in the present embodiment, which reads an image of a printed matter printed and output by the printing apparatus 110 and acquires an inspection image to be inspected. The image defect determination unit 172 compares the master image generated by the master image generation unit 170 with the inspection image acquired by the image reading unit 174, and determines the presence or absence of defects in the printed material.

  Hereinafter, the operation of the printing apparatus 110 will be described. In response to a print job execution instruction, the system control unit 112 extracts a print job received via the DFEI / F unit 120 or a print job stored in the HDD 126 as a RIP image, and sends it to the image processing control unit 128. Send. The mechanism control unit 132 issues an image signal output command to the printing control unit 130 based on the print command from the system control unit 112. The printing control unit 130 issues an image processing data output command to the image processing control unit 128 based on the image signal output command from the mechanism control unit 132. The image processing control unit 128 converts the RIP image into a print image in accordance with an image data output command from the printing control unit 130, transmits the converted print image to the printing control unit 130, and prints the printed matter for inspection. It transmits to the inspection device 150.

  Hereinafter, the operation of the printed matter inspection apparatus 150 will be described with reference to FIGS. 3 and 4. 3 and 4 are block diagrams illustrating the operation of the printed matter inspection apparatus according to the present embodiment. FIG. 3 shows the operation in the paper registration mode, which is a print inspection mode for registering paper, while FIG. 4 shows the operation in the normal mode, which is a print inspection mode for inspection.

  First, the operation in the paper registration mode will be described with reference to FIG. The system control unit 152 receives job management information from the printing apparatus 110 via the external I / F unit 158. The job management information is temporarily stored in the storage unit 176 in the system control unit 152. The system control unit 152 includes a job management information processing unit 178. The job management information processing unit 178 extracts post-processing information from the received job management information, and uses the extracted post-processing information as an external I / F unit. Via 158, it is transmitted to the post-processing device 90 at the subsequent stage of the printed matter inspection device 150. The job management information processing unit 178 further extracts print management information from the job management information, and uses the extracted print management information as a master image generation unit 170, an image defect determination unit 172, an image reading unit 174, and a mechanism control unit 168. Forward to.

  FIG. 5A shows the data structure of job management information. The job management information shown in FIG. 5A includes a print inspection mode, page identification information, printing surface information, paper information, paper size information, and post-processing information. As the print inspection mode, “paper registration mode” or “normal mode” is designated. The page identification information is an identifier of the print page, and the identifier is counted up for every output of one page from the time of power-on, for example. The printing surface information indicates whether the page is the upper surface or the lower surface. The paper information indicates the paper type, and the paper size indicates the paper size. The post-processing information includes information necessary for a post-processing device connected after the printed matter inspection device. Of the job management information shown in FIG. 5A, items of post-processing information are extracted as the post-processing information described above, and other items are extracted as print management information.

  The job management information (the print management information extracted from the job management information) includes the item of the print inspection mode as described above, and the printed matter inspection apparatus 150 differs in two ways depending on the print inspection mode in the job management information. Execute the process. When the print inspection mode in the job management information indicates “paper registration mode”, a color conversion table for converting from the CMYK color space to the RGB color space and a correction coefficient at the time of resolution conversion are generated for each paper. These are registered as paper correction information. On the other hand, when the print inspection mode in the job management information is “normal mode”, a master image is generated, and image inspection processing is performed based on the master image.

  With reference to FIG. 3 again, the description of the paper registration mode will be continued. In the paper registration mode, image data including a predetermined chart is printed. The image reading unit 174 reads an image at a predetermined resolution with the image reading device 52 on the upper surface of the printed matter output from the printing device 110 and the image reading device 54 on the lower surface, and obtains a read image. Based on the print surface information included in the received print management information, the image reading unit 174 combines the acquired read image with the print management information and transmits the combined image as read image data to the image defect determination unit 172. When the print inspection mode in the print management information is the paper registration mode, the image defect determination unit 172 transmits the read image data as it is to the system control unit 152 without performing defect determination processing on the received read image data.

  The system control unit 152 stores the received read image data in the storage unit 176, and acquires the received read image data and chart information stored in advance in the HDD 162. The chart information is prepared for each paper size, and its contents include the paper size, the coordinates of the alignment marks at the four corners of the paper, and the color information of each color patch to be printed. The system control unit 152 acquires sheet size chart information corresponding to the received read image data.

  Based on the read image data and the chart information described above, the color conversion table generation unit 182 generates a color conversion table, and the correction coefficient calculation unit 180 calculates a correction coefficient. The generated color conversion table and correction coefficient are also stored in the HDD 162 as paper correction information. In this way, assuming a print output from the printing apparatus 110 and image reading by the image reading unit 174, a color conversion table and a correction coefficient for generating a master image for comparison from the print image are prepared. .

  FIG. 5B shows the data structure of the paper correction information. The paper correction information shown in FIG. 5B includes paper information, paper size, paper size correction information, and a color conversion table. The paper information indicates the paper type, and the paper size indicates the paper size. The paper size correction information is a correction coefficient for resolution conversion calculated by the correction coefficient calculation unit 180, and the color conversion table is a color conversion table generated by the color conversion table generation unit 182.

  Next, the operation in the normal mode will be described with reference to FIG. When the print inspection mode in the transferred print management information indicates “normal mode”, the image reading unit 174 reads the upper or lower surface of the printed matter with the image reading devices 52 and 54 at a predetermined resolution and gradation. The inspection image is acquired, and is transmitted to the image defect determination unit 172 as inspection image data. As will be described later, the resolution at which the image reading unit 174 reads an image can select a resolution that is high enough to allow a good inspection even for an image that requires a high-resolution inspection. For example, in the inspection standard based on human sensitivity evaluation, if a relatively low resolution such as 200 dpi is sufficient, for example, 400 dpi is selected. The gradation at the time of image reading is not particularly limited, and an 8-bit gradation or the like is selected according to a request for inspection accuracy.

  When the print inspection mode indicates “normal mode”, the master image generation unit 170 receives the print management information, and appropriately sends the paper correction information corresponding to the paper information specified by the print management information to the system control unit 152. Via the HDD 162. The master image generation unit 170 receives the print image from the printing apparatus 110 and converts it into the same data format as the inspection image. The RIP-processed print image is small-value (for example, 2 bits) image data at a predetermined resolution (for example, 1200 dpi) in the CMYK color space. The master image generation unit 170 converts this into multi-value (for example, 8 bits) image data of a predetermined resolution (400 dpi) in the same RGB color space as that at the time of image reading by the image reading unit 174, to obtain a master image, The image is sent to the image defect determination unit 172.

  When the print inspection mode in the print management information indicates “normal mode”, the image defect determination unit 172 is based on the master image generated by the master image generation unit 170 and the inspection image acquired by the image reading unit 174. Then, the difference between the images is taken and a defect in the image is inspected. The master image and the inspection image have substantially the same data size, resolution, and color space. In the image defect determination processing, the entire master image and inspection image are divided into a plurality of blocks, image comparison is performed for each block, the difference between corresponding pixels is detected, and compared with a predetermined threshold value. Thus, a defect determination result is generated.

  When the defect inspection is completed, the image defect determination unit 172 transmits an inspection image, inspection result image data including an edge image and a difference image, which will be described later, and defect determination information to the system control unit 152. Here, the defect determination information is a set of the defect determination result and the threshold value used for the defect determination. The defect determination result includes the upper left coordinates of the area of the m × n pixel determined as the defect, the lower right Coordinates and difference values are included.

  The system control unit 152 stores defect determination information in the storage unit 176 and the HDD 162. The system control unit 152 includes a defect determination result processing unit 184. The defect determination result processing unit 184 generates information to be transmitted to the printing apparatus 110 and the post-processing apparatus 90, and prints via the external I / F unit 158. Transmit to device 110 and post-processing device 90. The system control unit 152 further includes a user interface (UI) display information generation unit 186. The UI display information generation unit 186 generates UI display information for notifying the user of the image inspection result, and the user I / I The information is displayed on the operation panel 56 via the F unit 160. The image inspection result to be notified to the user is not particularly limited, but includes information indicating whether or not a defect is found, and if a defect is found, an inspection image, an edge image, a difference image, and the like at that location are included. obtain.

  The system control unit 152 further includes a network user interface (NUI) display information generation unit 188. The NUI display information generation unit 188 receives a NUI display request from an external device such as a personal computer via the network I / F unit 164. The NUI display request is a request for display information for displaying an image inspection result on the display device of the external apparatus. In response to receiving the NUI display request, the NUI display information generation unit 188 reads the defect determination information stored in the storage unit 176, and generates NUI display information for notifying the user of the image inspection result. The data is transmitted to the request source via the network I / F unit 164. The requesting external device can perform screen display based on the received NUI display information.

  Hereinafter, a master image generation process for generating a master image from a print image will be described in detail with reference to FIG. FIG. 6 is a diagram illustrating detailed blocks of the master image generation unit 170 according to the present embodiment. FIG. 6 also shows the flow of the master image generation process.

  As shown in FIG. 6, the master image generation unit 170 includes a resolution conversion unit 200, a paper correction processing unit 202, a color conversion unit 204, and a filter processing unit 206. The master image generation unit 170 also includes an internal storage area, and holds sheet correction information (correction coefficient and color conversion table) for generating a master image in the internal storage area. The master image generation unit 170 compares the paper information stored in the storage area with the paper information included in the received print management information. If they are different, first, the master image generation unit 170 is designated by the print management information. The sheet correction information corresponding to the sheet information being acquired is acquired from the HDD 162 via the system control unit 152.

  When the appropriate sheet correction information is acquired, the master image generation unit 170 converts the print image into the same data format as the inspection image based on the sheet correction information. Here, as an example, consider a case where the print image is CMYK 1200 dpi low-value (for example, 2 bits) data and the inspection image is resolution 400 dpi multi-value (for example, 8 bits) data. The resolution conversion unit 200 converts the original print image from resolution 1200 dpi low-value data at the time of output to 400 dpi multi-value (for example, 8 bits) data substantially the same as the inspection image. In addition, the paper correction processing unit 202 performs fine correction of the applicable dot gain and color using the correction coefficient included in the paper correction information. The color conversion unit 204 converts the original print image from the CMYK color space at the time of output to the RGB color space that is the color space of the inspection image captured by the image reading unit 174.

  The color conversion may be performed using a conversion formula assuming that the paper correction information includes coefficient data of the conversion formula, and color conversion with a high degree of freedom is assumed if the paper correction information includes an LUT (Look Up Table). You may carry out using a table. Furthermore, it is possible to obtain a master image by appropriately applying a correction filter including blur and aberration derived from the imaging optical system included in the taken inspection image.

  Hereinafter, details of the image defect determination processing for inspecting defects in units of blocks using the master image and the inspection image as inputs will be described with reference to FIGS. 7 to 9.

  In the inspection by image processing, when the print image is performed at a relatively high resolution, it is ideal to prepare the inspection master image and the inspection image at the same resolution and detect the difference between them. However, inspection standards based on human sensitivity evaluation often do not require inspection resolution up to that point, and in order to improve the inspection processing speed, a sufficient resolution is required experimentally, and the resolution must be as low as possible. It is desirable to be able to inspect. However, the experimentally determined resolution for inspection is almost always determined on an average and a large scale, and in rare cases, a shortage of inspection may occur for a special printed document. For example, artistic line drawings not found in normal natural images and text images, registration marks for positioning necessary for bookbinding of printed materials, and the like.

  If such fine lines that do not appear so often in normal printing are imaged with insufficient resolution, a clear image of 1 pixel or a blurred image of 2 pixels may be obtained depending on the position on the image. To do. Due to misalignment between the master image and the inspection image, the two kinds of images as described above are generated, and if it is determined to be a defect, this results in excessive detection. On the other hand, if the inspection resolution of the entire image is increased in order to cope with an image that requires a high-resolution inspection that is infrequent, such as the thin line described above, the inspection processing load increases and the inspection processing time is increased. It will be difficult to improve and reduce costs.

  Therefore, in the printed matter inspection apparatus 150 according to the present embodiment, the master image and the inspection image are divided into a plurality of blocks, and image comparison is performed for each corresponding block. Further, the image characteristics of the individual blocks of the master image are determined. Is adopted, and the resolution of the block at the time of comparison is switched in accordance with the image characteristics. More specifically, the edge characteristics of each block of the master image are detected, and the resolution is reduced by temporarily reducing the resolution of the blocks with strong edge characteristics and the blocks with weak edge characteristics. Compare and match. This makes it possible to inspect at higher speed than inspecting everything with high resolution and with higher accuracy than inspecting everything with low resolution, and to improve inspection accuracy without increasing inspection time. .

  FIG. 7 is a detailed block diagram of the image defect determination unit 172 according to the present embodiment. FIG. 7 also shows the flow of image defect determination processing. As shown in FIG. 7, the image defect determination unit 172 includes block division units 210 and 212, an edge measurement and block distribution unit 214, a block distribution unit 216, reduction processing units 218 and 220, and a difference detection unit 222. It is comprised including.

  The block dividing unit 210 and the block dividing unit 212 are dividing units that divide the master image and the inspection image into predetermined area units, respectively. In the embodiment to be described, the predetermined area unit is a block unit of the number of pixels × the number of pixels from the viewpoint of simplifying the comparison operation.

  FIG. 8 is a diagram illustrating an aspect in which the entire image is divided into a plurality of blocks in the present embodiment. The master image and the inspection image prepared in the above-described flow have the same data size, resolution, and color space. For example, it is assumed that the master image and the inspection image are data of a resolution of 400 dpi and an RGB color space having a size of 1200 × 1800 pixels.

  Each of the block dividing unit 210 and the block dividing unit 212 divides a master image and an inspection image as shown in FIG. 8A into a plurality of blocks as shown in FIG. 8B. In the example shown in FIG. 8B, the size of one block is 150 × 100 pixels, and the entire 1200 × 1800 pixels described above is divided into a total of 144 (8 × 18 blocks) blocks. .

  Master images and inspection images are usually difficult to overlay exactly due to their imaging conditions (lens distortion, lens-to-paper distance variation, transport blur, paper expansion and contraction), etc. . However, it is possible to perform a good inspection by dividing into blocks and aligning each block by moving it minutely. In the embodiment to be described, it is assumed that the blocks are appropriately aligned by, for example, the existing technology disclosed in Patent Document 1 and the like, and detailed description is not given here. Note that the alignment marks at the four corners of the image can be used for ancestor adjustment in the first stage of overall alignment. Patent Document 1 can be referred to for such a multi-stage alignment technique.

  Referring to FIG. 7 again, the edge measurement and block distribution unit 214 measures the edge characteristics for each block of the master image, and distributes the processing flow of the blocks based on the measured edge characteristics. The edge measurement and block distribution unit 214 constitutes a measurement unit in this embodiment. More specifically, the edge measurement and block distribution unit 214 applies an edge detection filter to the block images constituting the master image, and generates an edge image for each block. Then, the edge characteristics are evaluated based on the sum and / or distribution of the portions corresponding to the edges in the edge image. Algorithms used for determining the sum and distribution of edge portions in a block can be combined by known techniques such as an edge detection filter, and are not particularly limited. Typical examples of the edge detection filter include a Sobel filter and a Prewitt filter.

  For blocks with strong edge characteristics, the edge measurement and block distribution unit 214 and the block distribution unit 216 allow the difference detection unit 222 to distribute the block image as it is. On the other hand, for a block with weak edge characteristics whose edge characteristics are less than the specified value, the edge measurement and block distribution unit 214 and the block distribution unit 216 first distribute block images to the reduction processing units 218 and 220.

  The reduction processing unit 218 and the reduction processing unit 220 reduce the input master image and the block image of the inspection image, respectively, with respect to the blocks whose edge characteristics are less than the specified value. For example, it is lower than the original resolution of 400 dpi, and can be lowered to a resolution of 200 dpi sufficient for the inspection standard based on human sensitivity evaluation.

  The difference detection unit 222 detects a defect by comparing and comparing the master image and the block image of the inspection image that are input at the original resolution as they are with respect to the block whose edge characteristics are not less than the standard. For blocks whose edge characteristics are less than the specified value, the difference detection unit 222 detects defects by comparing and collating the reduced block images of the master image and the inspection image. The master image and the inspection image are overlapped for each block, a difference between each pixel constituting the block is detected, and when the difference value of the pixel exceeds a threshold value, the portion is determined as a defect.

  FIG. 9 is a flowchart of image inspection processing executed by the printed matter inspection apparatus according to the present embodiment. The process shown in FIG. 9 is started from step S100. In step S101, the printed matter inspection apparatus 150 resets the block counter. Here, the block counter identifies a block to be processed among all the divided blocks, and the first block is selected by reset.

  In step S102, the printed matter inspection apparatus 150 acquires a master image block (hereinafter referred to as a master block image) and an inspection image block (hereinafter referred to as an inspection block image) to be processed. In step S103, the printed matter inspection apparatus 150 applies an edge detection filter to the master block image to generate an edge image. In step S104, the printed matter inspection apparatus 150 determines whether the block image has many edges based on the edge image. Here, based on the sum and distribution of the pixel values of the portion determined to be an edge, if the block has a high contrast between a flat portion and a portion with an edge, the character region or line drawing having a strong edge It is determined that the block is a region. On the other hand, if the block does not include many edges, it is determined as a photographic area, a natural image area, or a picture area having a weak edge.

  If it is determined in step S104 that the block image has many edges (YES), the process branches to step S105. In step S105, the difference between the master block image and the inspection block image is calculated with the original resolution (for example, 400 dpi), and the process proceeds to step S108. This is because when such a block detects a pixel difference at a low resolution, an error due to an edge shift increases, and as a result, a defect that can be tolerated by human judgment is excessively detected. Therefore, it is highly necessary to inspect with a relatively high resolution.

  In contrast, if it is determined in step S104 that the image is not a block image with many edges (NO), the process branches to step S106. In step S106, a batch reduction process is performed on the master block image and the inspection block image, and the resolution is lowered to a resolution lower than the original (for example, 200 dpi). In step S107, the difference between the master block image and the inspection block image is calculated at a resolution lower than the original (for example, 200 dpi), and the process proceeds to step S108. In such a block, since the change in the difference value between adjacent pixels is gradual, the difference value between the master image and the inspection image caused by the misregistration is increased even when the pixel difference detection is performed at a relatively low resolution. Is not excessive. Therefore, by inspecting at a relatively low resolution, the processing speed can be increased without reducing the detection accuracy.

  Since such batch reduction processing can be performed collectively for the entire block image, even if such extra processing is added, compared to the case of calculating the difference detection processing between high resolution images, The calculation load is small, and the overall speed can be increased. In particular, the reduction processing can be configured relatively simply with hardware, and is highly versatile. Furthermore, since dedicated hardware logic is also provided, it is possible to increase the operation speed at low cost by using such hardware.

  Referring to the image example shown in FIG. 8, a block in which the characters “Let's eat vegetables!” Have many edges. Therefore, empirically, a determination threshold value for determining whether or not a block image has many edges may be set so as to detect a difference at a high resolution for such a block image. On the other hand, the background portion and the pattern portion such as tomato and broccoli have relatively few edges, so if you set a decision threshold to perform difference detection processing at low resolution for such block images Good.

  When block difference detection ends in step S105 or step S107, the printed matter inspection apparatus 150 stores the block difference detection result in step S108, and advances the process to step S109. In step S109, the printed matter inspection apparatus 150 refers to the block counter and determines whether or not the processing for all the blocks has been completed. If it is determined in step S109 that the block counter has not reached the maximum value and there are still unprocessed blocks (NO), the block counter is incremented in step S110, and the process loops to step S101. Then, the next block is processed. On the other hand, if a predetermined number of blocks (144 in the example shown in FIG. 8) is performed and it is determined in step S109 that all blocks have been completed (YES), the process branches to step S111, and all blocks are The difference detection result is stored as a difference image detection result, and the image inspection process is terminated.

  According to the printed matter inspection apparatus 150 of the above-described embodiment, the master image and the inspection image are divided into a plurality of blocks, and further, the image characteristics of each block of the master image are detected, and according to the image characteristics, It is configured to perform image comparison for each block by switching the resolution of the block at the time of comparison. In a specific embodiment, image edge characteristics of individual blocks of the master image are detected, and the resolution is reduced by temporarily reducing the resolution of blocks having strong edge characteristics and the blocks having weak edge characteristics. Compared and verified.

  By adopting the above-described configuration, it is possible to inspect at a higher speed than inspecting everything at a high resolution and at a higher accuracy than inspecting everything at a low resolution, and without increasing the inspection time, the inspection accuracy is increased. Can be improved. In addition, since the edge characteristics of each block image of the master image are detected, the resolution can be switched according to the image frequency attribute of the actual block image, not the semantic attribute in the image. The overall processing amount can be reduced. In addition, the inspection image and master image are generated with a relatively high resolution, and the block image with a relatively low resolution is prepared by performing a reduction process in a batch for the weak edge portion, thereby suitably reducing the calculation load. can do.

  As described above, according to the present embodiment, in an image inspection process in which an output product from an output device is read and inspected, an image portion that rarely needs to be inspected at a high resolution is identified, and this is performed. It is possible to provide an image processing apparatus, an output object inspection method, and a program capable of improving the overall average processing speed while avoiding insufficient inspection in an image portion.

  The functional unit can be realized by a computer-executable program written in a legacy programming language such as assembler, C, C ++, C #, Java (registered trademark), an object-oriented programming language, or the like. ROM, EEPROM, EPROM , Stored in a device-readable recording medium such as a flash memory, a flexible disk, a CD-ROM, a CD-RW, a DVD-ROM, a DVD-RAM, a DVD-RW, a Blu-ray disc, an SD card, an MO, or through an electric communication line Can be distributed. In addition, a part or all of the functional unit can be mounted on a programmable device (PD) such as a field programmable gate array (FPGA) or mounted as an ASIC (application-specific integration). Circuit configuration data (bit stream data) downloaded to the PD in order to implement the above functional unit on the PD, HDL (Hardware Description Language) for generating the circuit configuration data, VHDL (VHSIC (Very High Speed) Integrated Circuits) Hardware Description Language)), data described in Verilog-HDL, etc. can be distributed by a recording medium.

  In the above-described embodiment, the printed matter inspection apparatuses 50 and 150 connected to the printing apparatus 20 are described as an example of the image processing apparatus, but are not particularly limited. In another embodiment, the image inspection is performed by receiving the input of the image reading data by the image reading device provided in the paper discharge unit of the printing device 20 and the RIP image or the print image input to the printing device 20. The image processing circuit may be configured as an information processing apparatus such as a computer, or may be configured as an image forming apparatus such as a printing apparatus provided with a printed matter inspection function in addition to a printing function. It may be configured.

  Although the embodiments of the present invention have been described so far, the embodiments of the present invention are not limited to the above-described embodiments, and those skilled in the art may conceive other embodiments, additions, modifications, deletions, and the like. It can be changed within the range that can be done, and any embodiment is included in the scope of the present invention as long as the effects of the present invention are exhibited.

DESCRIPTION OF SYMBOLS 10 ... Printing system, 12 ... Image generation controller, 20, 110 ... Printing apparatus, 22 ... Photosensitive drum, 24 ... Conveyance roller, 26 ... Conveyance roller, 28 ... Roller, 30 ... Intermediate transfer belt, 32 ... Feeding unit, 34 ... feed roller, 36 ... fixing roller, 38 ... reverse path, 40 ... operation panel, 50, 150 ... printed material inspection device, 52 ... image reading device, 54 ... image reading device, 56 ... operation panel, 90 ... post-processing Device ... 92 ... Tray 100 ... Hardware block 112 ... System control unit 114 ... CPU 116 ... Memory 118 ... User I / F unit 120 ... DFEI / F unit 122 ... Network I / F unit 124 ... External I / F unit, 126 ... HDD, 128 ... Image processing control unit, 130 ... Printing control unit, 132 ... Mechanism control unit, 152 ... System control unit, 154 CPU, 156, memory, 158, external I / F unit, 160, user I / F unit, 162, HDD, 164, network I / F unit, 168, mechanism control unit, 170, master image generation unit, 172, image Defect determination unit, 174 ... Image reading unit, 176 ... Storage unit, 178 ... Job management information processing unit, 180 ... Correction coefficient calculation unit, 182 ... Color conversion table generation unit, 184 ... Defect determination result processing unit, 186 ... UI display Information generation unit, 188... NUI display information generation unit, 200... Resolution conversion unit, 202... Paper correction processing unit, 204... Color conversion unit, 206 ... filter processing unit, 210 ... block division unit, 212. ... edge measurement and block distribution unit, 216 ... block distribution unit, 218 ... reduction processing unit, 220 ... reduction processing unit, 222 ... difference detection unit

JP 2013-186562 A Japanese Patent No. 5308735 Japanese Patent No. 4834387 JP 2013-046092 A

Claims (8)

Inspection image acquisition means for acquiring an inspection image generated by reading an output product from the output device at a predetermined resolution;
A reference image generating means for generating a reference image having substantially the same resolution as that of the inspection image, based on the original data of the output product;
A dividing unit that divides the reference image and the inspection image into regions, respectively.
Measuring means for measuring edge characteristics for each region unit of the reference image;
Reduction processing means for reducing the image of the reference image in units of regions and the image of the corresponding region of the inspection image, each of which has an edge characteristic that is less than a regulation;
An image processing apparatus comprising: a defect detection unit that detects a defect by comparing between the reference image and the reduced image of the inspection image with respect to a region unit in which the edge characteristic is less than a specified value.   The measuring means generates an edge image from an image of a region unit of the reference image, and evaluates the edge characteristics based on the sum and / or distribution of a portion corresponding to the edge in the edge image. The image processing apparatus according to 1.   The image processing apparatus according to claim 1, wherein the area unit is a block.   The image processing according to claim 3, wherein the defect detection unit detects a difference between each pixel constituting the block of the reference image and the inspection image, and detects a defect based on a detection result. apparatus. The reference image generation means includes
Resolution conversion means for converting an image based on the original data from a resolution at the time of output to substantially the same resolution as the inspection image;
The image processing apparatus according to claim 1, further comprising: a color conversion unit that converts an image based on the original data from a color space at the time of output to a color space for image reading. An output inspection method for inspecting an output from an output device, wherein a computer, an image processing circuit, or a combination thereof,
Obtaining an inspection image generated by reading the output product at a predetermined resolution; and
Generating a reference image having substantially the same resolution as the inspection image, based on the original data of the output product;
Dividing the reference image and the inspection image into regions, respectively.
Measuring edge characteristics for each region unit of the reference image;
Reducing the region-by-region image of the reference image and the corresponding region-by-region image of the inspection image, each having an edge characteristic that is less than a specified value;
A step of detecting a defect by comparing between the reference image and a reduced image of the inspection image with respect to a region unit in which the edge characteristic is less than a specified value. Measuring the edge characteristics comprises:
Generating an edge image from an image of a region unit of the reference image;
The output object inspection method according to claim 6, further comprising: evaluating the edge characteristics based on one or both of a sum total and a distribution of portions corresponding to edges in the edge image. Inspection image acquisition means for acquiring an inspection image generated by reading an output product from the output device at a predetermined resolution;
A reference image generating means for generating a reference image having substantially the same resolution as that of the inspection image based on the original data of the output product;
A dividing unit that divides the reference image and the inspection image into regions, respectively.
Measuring means for measuring edge characteristics for each region unit of the reference image;
Reducing processing means for reducing the image of the region unit of the reference image whose edge characteristic is less than the specification and the image of the corresponding region unit of the inspection image, respectively, and for the region unit whose edge property is less than the specification, the reference image and A program for causing a computer or a programmable device to function as defect detection means for detecting a defect by comparing the reduced images of the inspection image.