JP2015172506A - Image inspection device, image inspection system, and image inspection method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect a defect with a relatively long period in periodical defects with high accuracy in a system of inspecting an image output on a recording medium by image formation and output.SOLUTION: An image inspection system comprises: acquiring changes in a value relating to the density of a read image depending on a position along a conveyance direction of a recording medium on the basis of the read image; removing variations with a predetermined frequency or higher from periodical variations generating in the changes in the value relating to the density; detecting an inflection point where a direction of changes depending on a position in the conveyance direction of the recording medium varies in the changes in the value relating to the density of the image from which variations with a predetermined frequency or higher are removed; and determining the presence of defects periodically generating in the conveyance direction of the recording medium in the read image, on the basis of the detected inflection point.

Description

  The present invention relates to an image inspection apparatus, an image inspection system, and an image inspection method, and more particularly, to an increase in accuracy of periodic defect detection processing.

  Conventionally, inspection of printed matter has been performed manually, but in recent years, an apparatus for performing inspection has been used as post-processing of offset printing. In such an inspection apparatus, a master image serving as a reference is generated by manually selecting and reading a non-defective product from the read image of the printed matter, and a corresponding portion of the master image and the read image of the printed matter to be inspected. And the defect of the printed matter is discriminated by the degree of these differences.

  However, plateless printing devices such as electrophotography, which have become popular in recent years, are good at printing a small number of parts, and there are many cases where the content of printing on each page differs, such as variable printing, and a master image is generated from printed matter like an offset printer. In comparison, it is inefficient. In order to cope with this problem, it is conceivable to generate a master image from print data. Thereby, it is possible to efficiently cope with variable printing.

  In addition, as a method for detecting banding, which is a periodic defect appearing in an image, a difference image obtained by calculating a difference value between each pixel of a read image and a master image is generated, and pixels of each pixel constituting the difference image A method of performing frequency analysis on values has been proposed (see, for example, Patent Document 1). By such processing, even if the difference value is a small value and an error that is permitted by comparison with a simple threshold value, it is possible to determine that it is a defect by having periodicity. it can.

  However, according to the periodic error extraction process using frequency analysis, although banding in which a change in density appears periodically in one page can be detected, it is close to the length of one page in the sub-scanning direction. It is difficult to detect a banding of a period or a banding of a period longer than that.

  The present invention has been made in view of the above circumstances, and in a system that inspects an image output on a recording medium by image formation output, detection processing of a relatively long-period defect among periodic defects is enhanced. The purpose is to be accurate.

  In order to solve the above problems, an aspect of the present invention is an image inspection apparatus that inspects a read image obtained by reading an image formed and output on a recording medium, and the image formed and output is read. A read image acquisition unit that acquires the read image generated in this manner, and a density that acquires a change in the value related to the density of the read image according to the position in the direction in which the recording medium is conveyed based on the acquired read image A change acquisition unit, a concentration fluctuation processing unit that removes fluctuations greater than or equal to a predetermined frequency among periodic fluctuations occurring in a change in the value related to the density, and fluctuations greater than or equal to the predetermined frequency are removed. An inflection point detecting unit for detecting an inflection point, which is a point at which the direction of change according to the position in the direction in which the recording medium is conveyed changes with respect to a change in the value related to the density, Based on, characterized in that it comprises a determining periodic defect determining section that periodically occurring defects in the conveying direction of the reading the recording medium in the image occurs.

  According to the present invention, in a system for inspecting an image output on a recording medium by image formation output, a relatively long cycle defect detection process can be performed with high accuracy among periodic defects.

It is a figure showing composition of an image inspection system containing an inspection device concerning an embodiment of the present invention. It is a block diagram which shows the hardware constitutions of the test | inspection apparatus which concerns on embodiment of this invention. It is a block diagram which shows the function structure of DFE, an engine controller, a print engine, and an inspection apparatus which concern on embodiment of this invention. It is a figure which shows the aspect of the comparison test | inspection which concerns on embodiment of this invention. It is a figure which shows the mechanical structure of the print engine and inspection apparatus which concern on embodiment of this invention. It is a block diagram which shows the function structure of the master image process part which concerns on embodiment of this invention. It is a block diagram which shows the function structure of the test | inspection control part which concerns on embodiment of this invention. It is a flowchart which shows the image test | inspection operation | movement which concerns on embodiment of this invention. It is a figure which shows the example of the flat area | region which concerns on embodiment of this invention, and the area | region which is not so. It is a flowchart which shows the flat area | region determination operation | movement which concerns on embodiment of this invention. It is a figure which shows the parameter used in the flat area | region determination operation | movement which concerns on embodiment of this invention. It is a figure which shows the example of the image of flat area | region determination object which concerns on embodiment of this invention. It is a figure which shows progress of the flat area | region determination operation | movement which concerns on embodiment of this invention. It is a figure which shows progress of the flat area | region determination operation | movement which concerns on embodiment of this invention. It is a figure which shows the flat area | region determination result which concerns on embodiment of this invention. It is a flowchart which shows the banding determination operation | movement which concerns on embodiment of this invention. It is a figure which shows the average brightness according to the subscanning position which concerns on embodiment of this invention. It is a figure which shows the correlation table of the image size and observation distance which concerns on embodiment of this invention, and the correlation table of image size and the minimum detection width. It is a figure which shows the small pitch removal process result which concerns on embodiment of this invention. It is a figure which shows the data compression process result which concerns on embodiment of this invention. It is a flowchart which shows the inflection point detection process which concerns on embodiment of this invention. It is a figure which shows the inflection point detection process result which concerns on embodiment of this invention. It is a figure which shows the calculation aspect of the lightness gradient which concerns on embodiment of this invention. It is a figure which shows the correlation table of the standard deviation maximum value and flatness which concern on embodiment of this invention. It is a figure which shows the area | region of the banding determination process target which concerns on embodiment of this invention. It is a figure which shows the correlation table of the flatness and correction coefficient which concern on embodiment of this invention.

Embodiment 1 FIG.
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In this embodiment, a periodic defect occurring in an image is detected in an image inspection system including an inspection apparatus that inspects an output result by comparing a read image obtained by reading an output result of image formation output with a master image. How to do will be described. FIG. 1 is a diagram illustrating an overall configuration of an image forming system according to the present embodiment.

  As shown in FIG. 1, the image forming system according to the present embodiment includes a DFE (Digital Front End) 1, an engine controller 2, a print engine 3, an inspection device 4, and an interface terminal 5. The DFE 1 is an image processing apparatus that generates image data to be printed based on a received print job, that is, bitmap data that is an output target image, and outputs the generated bitmap data to the engine controller 2.

  The engine controller 2 controls the print engine 3 based on the bitmap data received from the DFE 1 to execute image formation output. Further, the engine controller 2 uses the bitmap data received from the DFE 1 as information serving as a basis of an inspection image to be referred to when the inspection device 4 inspects the result of image formation output by the print engine 3. Send to.

  The print engine 3 is an image forming apparatus that executes image formation output on a sheet as a recording medium based on bitmap data in accordance with control of the engine controller 2. As the recording medium, in addition to the above-described paper, a sheet-like material such as a film or plastic can be used as long as it is an object for image formation output.

  The inspection device 4 generates a master image based on the bitmap data input from the engine controller 2. The inspection device 4 is an image inspection device that inspects an output result by comparing a read image generated by reading a sheet output from the print engine 3 with a reading device with the generated master image.

  When the inspection device 4 determines that the output result is defective, the inspection device 4 notifies the engine controller 2 of information indicating the page determined as defective. Thereby, reprint control of the defective page is executed by the engine controller 2.

  In addition, the inspection apparatus 4 according to the present embodiment analyzes the difference image obtained by calculating the difference value between the read image and the master image described above, so that the density of the read image periodically changes in the sheet conveyance direction. The process for detecting the defect called is performed. This processing in banding detection is one of the gist of the present embodiment.

  The interface terminal 5 is an information processing terminal for displaying a GUI (Graphical User Interface) for confirming a defect determination result by the inspection apparatus 4 and a GUI for setting parameters in the inspection, and is a PC (Personal Computer). This is realized by a general information processing terminal such as.

  Here, hardware constituting the DFE 1, the engine controller 2, the print engine 3, the inspection apparatus 4, and the interface terminal 5 according to the present embodiment will be described with reference to FIG. FIG. 2 is a block diagram illustrating a hardware configuration of the inspection apparatus 4 according to the present embodiment. In FIG. 2, the hardware configuration of the inspection apparatus 4 is shown, but the same applies to other apparatuses.

  As shown in FIG. 2, the inspection apparatus 4 according to the present embodiment has the same configuration as an information processing apparatus such as a general PC (Personal Computer) or a server. That is, the inspection apparatus 4 according to the present embodiment includes a CPU (Central Processing Unit) 10, a RAM (Random Access Memory) 20, a ROM (Read Only Memory) 30, an HDD (Hard Disk Drive) 40, and an I / F 50. Connected through. Further, an LCD (Liquid Crystal Display) 60, an operation unit 70, and a dedicated device 80 are connected to the I / F 50.

  The CPU 10 is a calculation means and controls the operation of the entire inspection apparatus 4. The RAM 20 is a volatile storage medium capable of reading and writing information at high speed, and is used as a work area when the CPU 10 processes information. The ROM 30 is a read-only nonvolatile storage medium and stores a program such as firmware. The HDD 40 is a non-volatile storage medium that can read and write information, and stores an OS (Operating System), various control programs, application programs, and the like.

  The I / F 50 connects and controls the bus 90 and various hardware and networks. The LCD 60 is a visual user interface for the user to check the state of the inspection apparatus 4. The operation unit 70 is a user interface such as a keyboard and a mouse for the user to input information to the inspection apparatus 4.

  The dedicated device 80 is hardware for realizing a dedicated function in the engine controller 2, the print engine 3, and the inspection apparatus 4. In the case of the print engine 3, a transport mechanism that transports a sheet that is an image formation output target, A plotter device that executes image formation output on a paper surface. Further, the engine controller 2 and the inspection device 4 are dedicated arithmetic devices for performing image processing at high speed. Such an arithmetic unit is configured as an ASIC (Application Specific Integrated Circuit), for example. A reading device that reads an image output on a paper surface is also realized by the dedicated device 80.

  In such a hardware configuration, the software control unit is configured by the CPU 10 performing calculations in accordance with a program stored in the ROM 30 or a program read to the RAM 20 from a recording medium such as the HDD 40 or an optical disk (not shown). The A functional block that realizes the functions of the DFE 1, the engine controller 2, the print engine 3, the inspection apparatus 4, and the interface terminal 5 according to the present embodiment is configured by a combination of the software control unit configured as described above and hardware. Is done.

  FIG. 3 is a block diagram illustrating functional configurations of the DFE 1, the engine controller 2, the print engine 3, and the inspection device 4 according to the present embodiment. In FIG. 3, data transmission / reception is indicated by a solid line, and the flow of paper is indicated by a broken line. As illustrated in FIG. 3, the DFE 1 according to the present embodiment includes a job information processing unit 101 and a RIP processing unit 102. The engine controller 2 includes a data acquisition unit 201, an engine control unit 202, and a bitmap transmission unit 203. The print engine 3 includes a print processing unit 301. The inspection device 4 includes a reading device 400, a read image acquisition unit 401, a master image processing unit 402, an inspection control unit 403, and a comparative inspection unit 404.

  The job information processing unit 101 executes image formation output based on a print job input from outside the DFE 1 via a network or a print job generated based on image data stored in the DFE 1 by an operator's operation. Control. When executing the image formation output, the job information processing unit 101 causes the RIP processing unit 102 to generate bitmap data based on the image data included in the print job.

  The RIP processing unit 102 generates bitmap data for the print engine 3 to execute image formation output based on the image data included in the print job, under the control of the job information processing unit 101. Bitmap data is information of each pixel constituting an image to be imaged and output.

  The print engine 3 according to the present embodiment executes image formation output based on binary images of CMYK (Cyan, Magenta, Yellow, blackK). On the other hand, generally, image data included in a print job is a multi-valued image in which one pixel is expressed by multi-gradation such as 256 gradations. Therefore, the RIP processing unit 102 converts the image data included in the print job from a multi-valued image to a small-valued image, generates CMYK binary data for each color, and transmits it to the engine controller 2.

  The data acquisition unit 201 acquires bitmap data input from the DFE 1 and operates the engine control unit 202 and the bitmap transmission unit 203, respectively. The engine control unit 202 causes the print engine 3 to execute image formation output based on the bitmap data transferred from the data acquisition unit 201. The bitmap transmission unit 203 transmits the bitmap data acquired by the data acquisition unit 201 to the inspection apparatus 4 for generating a master image.

  The print processing unit 301 is an image forming unit that acquires bitmap data input from the engine controller 2, executes image formation output on printing paper, and outputs printed paper. The print processing unit 301 according to the present embodiment is realized by a general electrophotographic image forming mechanism, but other image forming mechanisms such as an ink jet method can also be used.

  The reading device 400 is an image reading unit that reads an image formed on a sheet of printing paper output by printing performed by the print processing unit 301 and outputs a read image. The reading device 400 is, for example, a line scanner installed in a conveyance path inside the inspection device 4 for printing paper output by the print processing unit 301. The scanning device 400 scans the paper surface of the printing paper to be conveyed on the paper surface. Read the formed image.

  The read image generated by the reading device 400 is an inspection target by the inspection device 4. Since the read image is an image generated by reading the paper surface of the paper output by the image forming output, the read image is an image indicating the output result. The read image acquisition unit 401 acquires information of a read image generated by reading the paper surface of the printing paper by the reading device 400. The information of the read image acquired by the read image acquisition unit 401 is input to the comparison inspection unit 404 for comparison inspection. Note that the input of the read image to the comparison inspection unit 404 is executed under the control of the inspection control unit 403. At that time, the inspection control unit 403 obtains the read image and inputs it to the comparison inspection unit 404.

  The master image processing unit 402 acquires the bitmap data input from the engine controller 2 as described above, and generates a master image that is an inspection image for comparison with the inspection target image. That is, the master image processing unit 402 functions as an inspection image generation unit that generates a master image, which is an inspection image for inspecting the read image, based on the output target image. The master image generation processing by the master image processing unit 402 will be described in detail later.

  The inspection control unit 403 is a control unit that controls the operation of the entire inspection apparatus 4, and each component included in the inspection apparatus 4 operates according to the control of the inspection control unit 403. The comparison inspection unit 404 compares the read image input from the read image acquisition unit 401 with the master image generated by the master image processing unit 402, and determines whether or not the intended image formation output is being executed. . The comparison inspection unit 404 is configured by the ASIC described above in order to quickly process a huge amount of calculation. In the present embodiment, the inspection control unit 403 functions as an image inspection unit by controlling the comparative inspection unit 404 and also functions as an inspection result acquisition unit that acquires an inspection result by the comparative inspection unit 404.

  In the comparison inspection unit 404, as described above, the 200 dpi read image and the master image expressed in 8 bits for each RGB color are compared for each corresponding pixel, and the difference value between the 8 bit pixel values for each RGB color described above for each pixel. Is calculated. Based on the magnitude relationship between the difference value thus calculated and the threshold value, the inspection control unit 403 determines the presence or absence of a defect in the read image. That is, the inspection control unit 403 functions as an image inspection unit by controlling each unit included in the inspection apparatus 4.

  When comparing the read image and the master image, the comparison inspection unit 404 superimposes the master image divided for each predetermined range on the read image corresponding to the divided range, as shown in FIG. The pixel value of the pixel, that is, the density difference is calculated. Such processing is realized by the inspection control unit 403 acquiring images in the overlapping range from the master image and the read image and inputting them to the comparison inspection unit 404.

  Further, the inspection control unit 403 shifts the position where the divided range is superimposed on the read image vertically, that is, while shifting the range of the image acquired from the read image vertically and horizontally, the total difference value calculated is calculated. The smallest position is determined as an accurate overlay position, and the difference value of each pixel calculated at that time is adopted as a comparison result. Therefore, the comparison inspection unit 404 can output the vertical and horizontal shift amounts when determined as the overlay position together with the difference value of each pixel.

  As shown in FIG. 4, each square divided into a grid is a predetermined range in which the difference values of the pixels described above are summed. In addition, the size of each division range illustrated in FIG. 4 is determined based on a range in which the comparison / inspection unit 404 configured by the ASIC can compare pixel values at a time as described above, for example.

  By such processing, the difference value is calculated after the read image and the master image are aligned. By comparing the difference value calculated in this way with a predetermined threshold value, a defect in the image is determined. Further, for example, even if there is a difference in scale between the entire read image and the entire master image, the influence of the scale is reduced by dividing and positioning for each range as shown in FIG. Is possible.

  Further, in each of the divided ranges as shown in FIG. 4, it is predicted that the amount of positional deviation between adjacent ranges is relatively close. Therefore, when performing the comparison inspection for each divided range, the calculation is performed while shifting the image in the vertical and horizontal directions by performing the above-described calculation while shifting the image in the vertical and horizontal directions with the positional deviation amount determined by the comparative inspection of the adjacent region as the center. Even if the number of times of performing is reduced, there is a high possibility that the calculation based on the accurate overlay position is executed, and the calculation amount as a whole can be reduced.

  Further, as described above, the inspection control unit 403 according to the present embodiment detects banding by analyzing the difference image. This function will be described in detail later.

  Next, the mechanical configuration of the print engine 3 and the inspection apparatus 4 and the paper conveyance path will be described with reference to FIG. As shown in FIG. 5, the print processing unit 301 included in the print engine 3 according to the present embodiment includes photosensitive drums 12 </ b> Y, 12 </ b> M, 12 </ b> C, and 12 </ b> K (hereinafter referred to as “photosensitive drums”) along the conveyance belt 11 that is an endless moving unit. In general, the photosensitive drum 12 is arranged in a row, and is called a so-called tandem type. That is, along the conveyance belt 11 that is an intermediate transfer belt on which an intermediate transfer image to be transferred to a sheet (an example of a recording medium) fed from the sheet feed tray 13 is formed, Photosensitive drums 12Y, 12M, 12C, and 12K are arranged in order from the upstream side.

  Each color image developed with toner on the surface of the photosensitive drum 12 for each color is superimposed on the conveyor belt 11 and transferred to form a full color image. The full-color image formed on the transport belt 11 in this way has a sheet surface of the paper transported on the path by the function of the transfer roller 14 at a position closest to the paper transport path indicated by a broken line in the drawing. Transcribed above.

  The paper on which the image is formed on the paper surface is further conveyed, the image is fixed by the fixing roller 15, and then conveyed to the inspection device 4. In the case of double-sided printing, the sheet on which an image is formed and fixed on one side is conveyed to the reversing path 16 and is reversed and conveyed to the transfer position of the transfer roller 14 again.

  The reading device 400 is configured by an information processing device inside the inspection device 4 by reading each surface of the paper conveyed from the print processing unit 301 in the paper conveyance path inside the inspection device 4 and generating a read image. The image is output to the read image acquisition unit 401. Further, the paper whose surface is read by the reading device 400 is further conveyed through the inside of the inspection device 4 and discharged to the paper discharge tray 410. 5 shows an example in which the reading device 400 is provided only on one side of the paper in the paper transport path in the inspection device 4, but in order to enable inspection of both sides of the paper, The reading device 400 may be arranged on each of both sides.

  As described above, since the print engine 3 includes a large number of driving parts including a plurality of rollers, vibrations are generated in the respective parts. When the vibration of a part that affects the image density is increased due to an assembly failure or the like, the image density periodically changes according to the vibration period, and the above-described banding occurs.

  For example, it is possible to detect a density change periodically generated in an image for one page by performing frequency analysis on the density change in the sub-scanning direction of the image for one page. However, if the period of density change is close to the length of one page in the sub-scanning direction or exceeds the length of one page in the sub-scanning direction, it is difficult to detect banding by frequency analysis. . The gist of the present embodiment is processing for detecting such long-period banding. Details will be described later.

  Next, a functional configuration of the master image processing unit 402 according to the present embodiment will be described. FIG. 6 is a block diagram illustrating an internal configuration of the master image processing unit 402. As illustrated in FIG. 6, the master image processing unit 402 includes a small-value / multi-value conversion processing unit 421, a resolution conversion processing unit 422, a color conversion processing unit 423, and an image output processing unit 424. Note that the master image processing unit 402 according to the present embodiment is realized by operating the dedicated device 80 described in FIG. 2, that is, hardware configured as an ASIC, according to software control.

  The low-value / multi-value conversion processing unit 421 generates a multi-value image by performing low-value / multi-value conversion processing on a binary image expressed in colored / colorless. The bitmap data according to the present embodiment is information to be input to the print engine 3, and the print engine executes image formation output based on CMYK (Cyan, Magenta, Yellow, blackK) color binary images. On the other hand, the read image, which is the image to be inspected, is a multi-valued image of RGB (Red, Green, Blue), which is the basic three primary colors, and is a multi-valued image. The image is converted into a multi-valued image. As the multivalued image, for example, an image expressed by 8 bits for each of CMYK can be used.

  The small value / multivalue conversion processing unit 421 performs an 8-bit extension process and a smoothing process as the small value / multivalue conversion process. The 8-bit extension process is a process of converting 0/1 data, which is 1 bit, into 8 bits, converting “0” to “0” and converting “1” to “255”. The smoothing process is a process of smoothing an image by applying a smoothing filter to 8-bit data.

  In this embodiment, the print engine 3 executes image formation output based on CMYK binary images, and the master image processing unit 402 includes a low-value multi-value conversion processing unit 421. Is an example. That is, when the print engine 3 executes image formation output based on a multi-value image, the low-value multi-value conversion processing unit 421 can be omitted.

  In addition, the print engine 3 may have a function of performing image formation output based on a low-value image such as 2 bits instead of 1 bit. In this case, it can be dealt with by changing the function of the 8-bit extension processing. That is, in the case of 2 bits, the gradation value is four values of 0, 1, 2, and 3. Therefore, in the case of 8-bit expansion, “0” is converted to “0”, “1” is converted to “85”, “2” is converted to “170”, and “3” is converted to “255”.

  The resolution conversion processing unit 422 performs resolution conversion so that the resolution of the multi-value image generated by the small-value multi-value conversion processing unit 421 matches the resolution of the read image that is the image to be inspected. In the present embodiment, since the reading device 400 generates a 200 dpi read image, the resolution conversion processing unit 422 converts the resolution of the multi-valued image generated by the small-value multi-value conversion processing unit 421 to 200 dpi. In addition, the resolution conversion processing unit 422 according to the present embodiment determines the size of the image after resolution conversion based on a predetermined magnification in consideration of the shrinkage of the paper output by the print processing unit 301 during resolution conversion. adjust.

  The color conversion processing unit 423 acquires an image whose resolution has been converted by the resolution conversion processing unit 422, and performs gradation conversion and color expression format conversion. The tone conversion processing is color tone conversion processing for matching the color tone of the master image with the color tone of the image formed on the paper surface by the print processing unit 301 and the color tone of the image read and generated by the reading device 400. .

  Such processing includes, for example, an image including color patches of various gradation colors formed on the paper surface by the print processing unit 301, and the gradation value of each color patch in the read image generated by reading the paper. This is performed using a gradation conversion table in which the gradation values in the original image for forming each color patch are associated with each other. That is, the color conversion processing unit 423 converts the gradation value of each color of the image output from the resolution conversion processing unit 422 based on such a gradation conversion table.

  The color representation format conversion process is a process for converting an image in the CMYK format into an image in the RGB format. As described above, since the read image according to this embodiment is an RGB format image, the color conversion processing unit 423 converts the CMYK format image that has been subjected to the gradation conversion processing into an RGB format. This color representation format conversion process is executed in the same manner as the above-described gradation conversion process, except that it is executed using a calculation formula for calculating the values of each color in the RGB format based on the values of each color in the CMYK format. , Based on a conversion table in which CMYK format values and RGB format values are associated with each other.

  Note that the above-described gradation conversion and color expression format conversion are performed by using a conversion table in which CMYK format values and RGB format values are associated with each other and considering the above-described gradation conversion. It is also possible to execute them simultaneously. Such processing can reduce the processing load.

  Through the processing up to the color conversion processing unit 423, a 200 dpi multi-value image expressed by RGB each color 8 bits (24 bits in total) is generated for each pixel. That is, in the present embodiment, the small-value / multi-value conversion processing unit 421, the resolution conversion processing unit 422, and the color conversion processing unit 423 function as an inspection image generation unit.

  The image output processing unit 424 outputs the master image generated by the processing up to the color conversion processing unit 423. As a result, the inspection control unit 403 acquires a master image from the master image processing unit 402.

  Next, a functional configuration of the inspection control unit 403 according to the present embodiment will be described. FIG. 7 is a block diagram illustrating a functional configuration of the inspection control unit 403 according to the present embodiment. FIG. 8 is a flowchart showing an image inspection operation for one page by the inspection control unit 403 according to the present embodiment. As shown in FIG. 7, the inspection control unit 403 according to the present embodiment includes an information input unit 431, a difference image acquisition unit 432, a defect determination unit 433, a flat region determination unit 434, a banding determination unit 435, and a controller communication unit 436. Including.

  In the inspection control unit 403 according to the present embodiment, as illustrated in FIG. 8, first, the information input unit 431 acquires a master image from the master image processing unit 402 (S801), and acquires a read image from the read image acquisition unit 401. Obtain (S802). Since the process of S801 and the process of S802 are not restricted in context, they may be executed in the reverse order or may be executed in parallel.

  As described with reference to FIG. 4, the information input unit 431 that has acquired the master image and the read image extracts a predetermined range of images from the master image and the read image and inputs the images to the comparison inspection unit 404, thereby comparing the inspection unit. In step S <b> 803, the image is subjected to a comparison inspection 404.

  By the processing of S803, a difference image indicating a difference value between each pixel constituting the read image and each pixel constituting the master image is generated, and the difference image acquisition unit 432 acquires the generated difference image. Defect determination is performed by the defect determination unit 433 based on the difference image acquired by the difference image acquisition unit 432 (S804).

  In step S804, the defect determination unit 433 determines whether each page includes a defect by comparing the value of each pixel constituting the difference image with a preset threshold value. In addition, by performing a labeling process or the like of the pixel determined as a defect, information indicating a defect determination result is generated by specifying a defect position, specifying a defect type, or the like.

  The flat area determination unit 434 determines a flat area in the image to be inspected based on the master image input from the master image processing unit 402 (S805). This flat area is an area where there is little change in the density of the image, for example, a plain area where the developer is not transferred, or a solid area. Details of the flat area determination processing by the flat area determination unit 434 will be described later. Note that the processing from S802 to S804 and the processing of S805 may be executed in the reverse order, or may be executed in parallel.

  Further, the banding determination unit 435 acquires a difference image from the difference image acquisition unit 432 and performs banding determination (S806). The banding determination process in S806 will be described in detail later. If the process of S805 is completed before the process of S806, the process of S804 and the process of S806 may be executed in the reverse order or may be executed in parallel.

  When the defect determination process and the banding determination process are completed, the controller communication unit 436 executes engine control such as a reprint request and a banding detection notification based on the defect determination result and the banding determination result (S807). By such processing, the image inspection operation according to the present embodiment is completed.

  Next, the flat region determination operation in S805 according to the present embodiment will be described. First, typical examples of an image area corresponding to a flat area and an image area not corresponding to the flat area will be described. FIG. 9 is a diagram showing an example of the pixel value, that is, the density of each pixel obtained by enlarging a part of each region in the case where the image of one page includes a photographic region and a plain region, by the number of hatched lines.

  As shown in FIG. 9, in the case of a photographic region, when a collection of partial pixels is seen, the density changes for each pixel. Here, in the master image generation processing described with reference to FIG. 6, the small value / multivalue conversion processing by the small value multivalue conversion processing unit 421 is performed. This process is a process for restoring a multi-valued image by applying a filter to a small-valued image. However, even if the process using the filter can restore the image as a whole to a similar state, FIG. It is difficult to accurately restore the shading in pixel units as shown in FIG.

  For this reason, when the difference value between the read image and the master image for a region with frequent shading changes as shown in FIG. 9 is calculated, a certain amount of difference value is calculated as an error even if no defect has occurred. End up. Even if banding determination is performed for such a difference image, there are many noises due to errors, and periodic defects cannot be accurately detected. Such an image region is an image region that does not correspond to a flat region.

  On the other hand, in the case of a plain region, when a collection of partial pixels is seen, it is uniformly white. For such a region, there is little error even if the difference value between the read image and the master image is calculated. Therefore, the region is suitable as a target for periodically detecting defects by banding determination on the difference value. In addition to the plain region, a solid coating region, a region where the density change is gentle, and the like have similar characteristics. Such an image region is an image region corresponding to a flat region, and extracting such a region is the purpose of the determination operation of the flat region in S805.

  FIG. 10 is a flowchart showing the flat region determination operation in S805 according to the present embodiment. As a premise for executing the operation shown in FIG. 10, the flat region determination unit 434 according to the present embodiment holds parameters as shown in FIG. 11. The unit detection size W is a pixel size of an area referred to in one flat area determination (hereinafter referred to as “flat determination area”), and is 80 pixels square in this embodiment.

  The minimum determination width H is a width indicating a minimum area as an area for performing banding determination among areas generated by connecting areas determined as flat areas. Periodic defects are generated, for example, due to the influence of a rotating member such as a roller for conveying paper or a photosensitive drum, and therefore basically periodically occur in the sub-scanning direction. Accordingly, an area having a narrow width in the sub-scanning direction is excluded from the determination minimum width H because it is less effective as a banding determination target. The minimum determination width H in this embodiment is 256 pixels.

  The standard deviation threshold TH is a threshold applied to the standard deviation calculated for the pixel value of one flat determination region, and when the calculated standard deviation, that is, the variation of the pixel value is equal to or smaller than the standard deviation threshold TH, It is determined that the area is a flat area. The standard deviation threshold TH in this embodiment is 2 on the premise of 256 gradations of RGB each color 8 bits.

  Based on the parameters as shown in FIG. 11, the flat area determination unit 434 refers to pixels for the flat determination area in order from the end of the master image to be determined as shown in FIG. 10 (S1001), and refers to the reference area. For all of the pixels, a standard deviation of pixel values is calculated for each color of RGB (S1002). The value of this standard deviation is used as the flatness indicating the change in shading for each pixel in each flatness determination area.

  Then, the flat area determination unit 434 compares the largest value among the calculated standard deviations of the RGB colors with the standard deviation threshold TH described in FIG. 11 (S1003), and if the calculated result is equal to or smaller than the standard deviation threshold TH. (S1003 / YES), assuming that the flatness determination area being referred to is a flat area, coordinates indicating the area being referred to are registered in the storage medium (S1004). This coordinate is, for example, the upper left coordinate of the area being referred to.

  When the calculation result is larger than the standard deviation threshold TH (S1003 / NO), or when the processing of S1004 is completed, the flat area determination unit 434 confirms whether the determination is completed for all areas of the determination target master image ( If not completed (S1005), the target region to be referred to as the flatness determination region is moved (S1008), and the processing from S1001 is repeated. In step S1008, the unit detection size W is moved by half of one side of the pixel in the sub-scanning direction. Further, when it reaches the end in the sub-scanning direction, it moves in the main scanning direction by the size of one side of the unit detection size W, and uses the head in the sub-scanning direction as a reference area.

  FIG. 12 is a diagram illustrating a part of a master image that is a target of flat area determination. In the example of FIG. 12, an image on which characters and figures are displayed with a solid paint area indicated by a diagonal line as a background is targeted. On the other hand, FIG. 13 is a diagram illustrating a state in which the processing up to S1004 has been completed for all regions. In FIG. 13, an area determined to be a flat area is surrounded by a dotted line.

  As shown in FIG. 13, when the determination is completed for all regions of the master image to be determined (S1005 / YES), the flat region determination unit 434 determines each unit detection size region determined as a flat region in the sub-scanning direction. (S1006). FIG. 14 is a diagram illustrating the processing result of S1006. In the processing of S1006, for example, regions where the main scanning positions of the coordinates registered in S1004 are the same and the sub-scanning position difference is the unit detection size W are integrated.

  As shown in FIG. 14, the flat region determination unit 434 that combines regions in the sub-scanning direction refers to the determination minimum width H described in FIG. 11, and the combined sub-scan width of each region is the determination minimum width H. The area that is less than the area is deleted from the flat area (S1007). FIG. 15 is a diagram illustrating a result of performing the processing of S1007 on the state of FIG. As shown in FIG. 15, an area that does not have a sufficient width in the sub-scanning direction as an area for detecting a periodic defect is excluded from the banding determination target area. By such processing, the flat region determination operation according to the present embodiment is completed.

  Note that, by the processing shown in FIG. 10, an area having a small change in shading for each pixel and having a sufficient width in the sub-scanning direction is extracted, but a solid area, that is, an area where the developer is not transferred. Alternatively, a process of excluding a high-density solid paint area close to black may be performed. In the plain area, the developer is not transferred, so the possibility of periodic defects is low, and in the high density solid-coated area, even if the periodic defects occur, the defects are inconspicuous because of the high density. This is because the merit of performing the banding determination is small.

  In such a process, for example, as the condition of “YES” in S1003 in FIG. 10, an RGB value in the flatness determination area, that is, an average value of pixel values is obtained, and the average value is equal to or less than a threshold value for determining a plain area. In addition, it can be added by adding that the threshold value is equal to or higher than a threshold value for determining a solid area having a high density. In the present embodiment, on the premise of 256 gradations for each color of RGB, the threshold value for determining the plain area is 220, for example, and the threshold value for determining the high-density solid paint area is 10, for example.

  Next, the banding determination operation in S806 of FIG. 8 according to the present embodiment will be described. FIG. 16 is a flowchart showing details of the operation in the banding determination process of S806. As shown in FIG. 16, the banding determination unit 435 converts the value of the difference image input from the difference image acquisition unit 432 into a YCbCr value (S1601). Since the lightness value Y has the greatest influence on human vision, the analysis is performed based on the lightness value Y in the banding determination process. Therefore, in the process of S1601, only Y among YCbCr may be obtained.

  Next, the banding determination unit 435 selects one of the flat regions connected in the sub-scanning direction determined as described with reference to FIG. 15, and the main scanning row, that is, the sub-scanning position is the same. The average value of the Y values of each pixel is calculated for each pixel (S1602). In the banding determination according to the present embodiment, the banding determination is performed in the sub-scanning direction in order to detect periodic defects that periodically occur in the sub-scanning direction. One value is obtained for each main scanning row by averaging so that the row of banding determination target values becomes one row. In other words, the banding determination unit 435 functions as a density change acquisition unit that acquires a change in the brightness value Y corresponding to the position in the direction in which the recording medium is conveyed, that is, a value related to the density of the difference image.

  FIG. 17A is a graph showing the lightness value for each sub-scanning position calculated by the process of S1602. Referring to FIG. 17 (a), it can be seen that the whole is rising to the right. This is considered to be a change in the original brightness value, but in an image forming apparatus such as a laser printer, an error may occur in an image formed by the main scanning position and the sub-scanning position. It may appear as an error of the overall value as shown in (a).

  In order to correct the error, the banding determination unit 435 performs inclination correction (1603). A general method can be used for the inclination correction in S1603. For example, the banding determination unit 435 calculates a correction angle so that the straight line obtained by the least square method is horizontal, and the angle is corrected. As described above, the average brightness at each sub-scanning position is corrected. FIG. 17B is a graph showing the average brightness with respect to the sub-scanning position after inclination correction.

  The banding determination unit 435 removes the small pitch from the average brightness of the sub-scanning position after the inclination correction is performed in this way (S1604). The removal of a small pitch is removal of fluctuations of a predetermined frequency or more among fluctuations occurring in the change in average brightness. That is, the banding determination unit 435 functions as a density fluctuation processing unit that removes fluctuations equal to or higher than a predetermined frequency among fluctuations occurring in the change in the value related to the average brightness, that is, density.

  In step S <b> 1604, the banding determination unit 435 sets the predetermined minimum detection width P (mm) to the filter size (a predetermined range at a position on the image) with respect to the average brightness of the sub-scanning position after the inclination correction. The moving average filtering process is performed.

  In the present embodiment, the minimum detection width P is determined according to an observation distance L that is a distance from the viewpoint when viewing the target image to the target image. The observation distance L can be determined by the size of the image. The observation distance L when a human visually observes images of general sizes such as A4 and B5 can be calculated based on the visual perception when the human observes the object, as shown in FIG. As described above, it is possible to generate a table in which “image size” such as “A0”, “A1”, and “A2” and the “observation distance L” corresponding thereto are associated with each other. In the table shown in FIG. 18A, the “observation distance L” increases as the “image size” increases.

  The minimum detection width P is calculated according to the image size based on the observation distance L according to such an image size. Specifically, the minimum detection width P is assumed to be a reciprocal of the value obtained by converting the spatial frequency 1 cycle / mm to cycle / mm with the minimum detection frequency. That is, the minimum detection width P is obtained by the following formula (1) using the observation distance L.

  P = L * π / 1 * 180 (1)

  By applying each observation distance L of the table shown in FIG. 18A to the above equation (1), an “image size” as shown in FIG. 18B and a “minimum detection width P” corresponding thereto. Can be generated as a table in which That is, in the table shown in FIG. 18B, the “minimum detection width P” increases as the “image size” increases.

  FIG. 19 is a graph in which the original graph before the small pitch is removed by the processing of S1604 is indicated by a dotted line, and the result of the small pitch being removed by the processing of S1604 is indicated by a solid line. The banding determination unit 435 performs data compression processing on the average brightness from which the small pitch has been removed in this way (S1605). In step S <b> 1605, the banding determination unit 435 divides the sub-scanning position into a predetermined range, and generates brightness value data in which the average value of the average brightness for each divided range is a representative value of the range.

  FIG. 20 is a diagram illustrating values obtained as a result of data compression performed on the graph indicated by the solid line in FIG. 19 by the processing of S1605, and the values of the points indicated by diamonds are the average brightness for each divided range. Is the average value. In the graph shown in FIG. 20, the scale of the average brightness axis is changed in accordance with the maximum and minimum values of the average brightness of the graph shown by the solid line in FIG.

  The banding determination unit 435 is an inflection that is a point where the direction of change in brightness changes from the points indicating the respective values calculated by the data compression in this way (points indicated by diamonds in FIG. 20). A point is detected (S1606). That is, the banding determination unit 435 functions as an inflection point detection unit that detects an inflection point. FIG. 21 is a flowchart illustrating an inflection point detection process in S1606. As illustrated in FIG. 21, the banding determination unit 435, for example, in order from the left of the graph illustrated in FIG. 20, the values of the points before and after the target point (hereinafter referred to as “target point”) are adjacent. Compare (lightness). That is, the points from the second point from the left to the second point from the right are set as target points in order.

  When the value of the target point is larger than the values of the adjacent points (S2101 / YES), the banding determination unit 435 determines an inflection point candidate that makes the target point in a convex state (hereinafter referred to as a “change in a convex state”). The “musical point” is detected as “mountain inflection point”) (S2102). When the detected inflection point candidate and the adjacent inflection point candidate are mountains (S2103 / YES), the banding determination unit 435 that has detected a mountain inflection point candidate is between the inflection point candidates between mountains. Among these points, the point having the minimum value is detected as an inflection point candidate that becomes concave (hereinafter, “the inflection point that becomes concave” is referred to as “the inflection point of the valley”) (S2104). . Note that when there are a plurality of minimum value points, the banding determination unit 435 detects any one point (for example, a point close to the center between inflection point candidates between mountains) as a valley inflection point candidate.

  On the other hand, if the value of the target point is smaller than the values of the adjacent points (S2101 / NO, S2105 / YES), the banding determination unit 435 detects the target point as a valley inflection point candidate (S2106). ). When the detected inflection point candidate and the adjacent inflection point candidate are valleys (S2107 / YES), the banding determination unit 435 that has detected the valley inflection point candidates is between the inflection point candidates between the valleys. Among these points, the maximum value point is detected as a mountain inflection point candidate (S2108). When there are a plurality of maximum value points, the banding determination unit 435 detects any one point (for example, a point close to the center between inflection point candidates between valleys) as a mountain inflection point candidate.

  On the other hand, if the value of the target point is greater than or equal to the value of one of the adjacent points, or less than or equal to the value of the other point (S2101 / NO, S2015 / NO), the banding determination unit 435 determines the target point. Not an inflection point candidate. Until the inflection point candidate detection processing for all points is completed (S2109 / YES), the banding determination unit 435 repeats the above-described processing with respect to points where the processing is not completed (S2109 / NO).

  When the inflection point candidate detection process for all points is completed (S2109 / YES), the banding determination unit 435 compares the difference (lightness difference) between adjacent inflection point candidate values with a predetermined threshold ( S2110). If the brightness difference between adjacent inflection point candidates is equal to or greater than a predetermined threshold, the banding determination unit 435 detects these inflection point candidates as inflection points (S2111). The banding determination unit 435 repeats the comparison process (S2112 / NO) until the comparison process between the brightness difference of all adjacent inflection point candidates and the threshold value is completed (S2112 / YES).

  FIG. 22 is a diagram showing a graph including inflection points detected from the points of the respective values calculated by the data compression shown in FIG. 20 by the process of S1606. The banding determination unit 435 calculates a lightness gradient V that is a change amount of a value related to the density between adjacent inflection points detected in this way (S1607). For example, as shown in FIG. 23, the coordinates where the third inflection point is located from the left are (X1, Y1), and the coordinates where the fourth inflection point is located from the left are (X2, Y2). Then, the banding determination unit 435 calculates the lightness gradient V between these inflection points by the following equation (2).

  V = (Y1-Y2) / (X1-X2) (2)

  The leftmost inflection point is calculated as a lightness gradient with respect to the first value of the graph (the value indicated by the square in FIG. 23), and the rightmost inflection point is the last value of the graph (FIG. 23, the brightness gradient with respect to the value of the point indicated by a square) is calculated.

  The banding determination unit 435 divides the sum of the absolute values of the lightness gradients V between all adjacent inflection points calculated in this way by the sub-scanning width (mm) of the determination region in accordance with the average lightness. (S1608). Here, the average brightness is the average value of the brightness values Y converted in S1601 in the banding determination target range. By correcting with the average brightness in this way, it is possible to eliminate the influence of the brightness of the original image. In the processing of S1608, the banding determination unit 435 obtains a coefficient by the ratio between the reference brightness and the above-described average brightness, and multiplies the coefficient by the coefficient.

  The value calculated in this way is a value indicating the strength of a periodic defect that is easily recognized by human vision in the banding determination target range. The banding determination unit 435 determines whether or not a periodic defect has occurred in the target region by comparing the value with a threshold provided for determining a periodic defect (S1609). . That is, the banding determination unit 435 functions as a periodic defect determination unit that determines whether or not a periodic defect has occurred. Note that the threshold for determining a periodic defect can be determined experimentally based on the analysis results of an image in which a periodic defect has occurred and an image in which no defect has occurred.

  The banding determination unit 435 repeatedly performs such processing on all the extracted regions as shown in FIG. 15 (S1610 / NO), and when completed for all the regions (S1610 / YES), the banding determination Complete the operation. If the banding determination unit 435 determines that a defect has occurred in S1609, the banding determination unit 435 outputs the determination result to the controller communication unit 436. The determination result includes coordinate information indicating an area determined to have a periodic defect, in addition to information indicating that a periodic defect has occurred.

  As described above, in the inspection apparatus 4 according to the present embodiment, the banding determination unit 435 removes the small pitch of the average brightness of the region determined to be a flat region by the flat region determination unit 434 and sets an inflection point. Detection is performed, and periodic defect detection processing is performed based on the brightness gradient between the detected adjacent inflection points. As a result, among the periodic defects that occur in the image, a relatively long-period defect detection process that exceeds the sheet length in the conveyance direction for one page, which is difficult to detect by frequency analysis, is performed with high accuracy. Can do.

In the above embodiment, the case where the area determined to be a flat area is the target of the above-described processing has been described as an example. In such a mode, for example, banding determination can be performed except for an image having a large influence of density fluctuation of the image itself such as a natural image such as a photograph, and periodic defect detection processing can be performed with higher accuracy. Can do.
However, it is not essential to process only the area determined to be a flat area, and a difference image between the entire read image and the entire master image may be processed. For example, when the text information is the central document information, only the density variation due to the text information is difficult to be affected by the banding determination. Therefore, even if the difference image between the entire scanned image and the entire master image is processed. Periodic defect detection processing can be performed with high accuracy.

  Moreover, in the said embodiment, the case where the inflection point was detected based on the difference image was demonstrated as an example. Since such an aspect is an image composed only of the difference value between the master image and the read image, it is possible to perform highly accurate banding determination without being affected by the density fluctuation in the original image. In addition, the same processing may be executed based on the read image. For example, when the density variation in the original image is small, it is possible to perform highly accurate banding determination without using a difference image. In addition, for example, the same processing may be executed based on the read image of the flat region with reference to the determination result of the flat region. In the case of a flat region, it is possible to perform highly accurate banding determination without being affected by density fluctuations in the original image.

  Further, in the above embodiment, the value calculated based on the sum of absolute values of the lightness gradient V between adjacent inflection points is compared with a threshold provided for determining periodic defects, thereby The case where it is determined whether or not a periodic defect has occurred in the region has been described as an example. In addition, the brightness difference threshold value in S2110 is set as a threshold value for determining a periodic defect, and when an adjacent inflection point having a brightness difference equal to or greater than the brightness difference threshold value is detected during the process in S2110, You may determine with the periodic defect having arisen. If the brightness difference between adjacent inflection points, that is, the amount of change in the value related to the density between adjacent inflection points is equal to or greater than a predetermined threshold, it can be determined that banding has occurred. In this case, it is not necessary to calculate the brightness gradient and perform processing for subsequent periodic defect determination.

Embodiment 2. FIG.
In the first embodiment, the case where periodic defect detection processing by banding determination is performed only on an area determined to be a flat area has been described as an example. However, this is an example, and banding determination may be performed on the entire difference image obtained by calculating the difference value between the read image and the master image, and weighting may be performed according to the flatness of the image to be inspected. . Hereinafter, such an aspect will be described.

  The inspection apparatus 4 according to the present embodiment has the same configuration as the aspect described in FIGS. 3 and 7 of the first embodiment. Further, in the image comparison inspection, an operation substantially similar to the operation described in FIG. 8 is performed. Here, the flat area determination unit 434 according to the present embodiment does not perform the flat area determination operation described in FIG. 10 in S805, but determines the unit detection size W square for all areas of the master image. A flatness calculation process is performed for each region.

  When calculating the flatness, the flat area determination unit 434 performs standard deviation calculation processing similar to S1002 of FIG. 10, for example, and determines the flatness based on the maximum value of the standard deviation calculated for each color of RGB. In determining the flatness, for example, a table associating the range of the standard deviation maximum value with the flatness as shown in FIG. 24 can be referred to. In the example of FIG. 24, a table is exemplified in which the value increases as the standard deviation maximum value increases, that is, the image is not flat.

  The flatness for each determination area calculated in this way is used in the banding determination process described with reference to FIG. As described above, the banding determination unit 435 according to the present embodiment performs banding determination on the entire range of the difference image. At this time, as shown in FIG. 25, the banding determination unit 435 according to the present embodiment performs FIG. 16 for each region that covers the entire range in the sub-scanning direction having a width corresponding to the unit detection size W in the main scanning direction. The banding determination process described in the above is performed.

  Further, the banding determination unit 435 according to the present embodiment performs a difference corresponding to the flatness of each determination region with respect to the value of the difference image input from the information input unit 431 before performing the process of S1601 in FIG. A value correction process is performed. The banding determination unit 435 according to the present embodiment performs correction processing by multiplying the difference value of each pixel by a coefficient corresponding to the flatness as shown in FIG.

  As a result, the region with low flatness is corrected so that the difference value of each pixel is small, so that the influence on the banding determination is also small. As a result, in a region where the flatness of the image is low, that is, a region where a change in shading for each pixel is large, it is possible to avoid erroneous detection of a periodic defect due to a difference value calculation error.

  In the present embodiment as well, a process of excluding a plain region and a high-density solid paint region close to black may be performed. As a result, it is possible to obtain the same effect as in the first embodiment.

1 DFE
2 Engine Controller 3 Print Engine 4 Inspection Device 5 Interface Terminal 10 CPU
20 RAM
30 ROM
40 HDD
50 I / F
60 LCD
70 Operation Unit 80 Dedicated Device 90 Bus 11 Conveyor Belt 12, 12Y, 10M, 12C, 12K Photosensitive Drum 13 Paper Tray 14 Transfer Roller 15 Fixing Roller 16 Reverse Pass 101 Job Information Processing Unit 102 RIP Processing Unit 201 Data Acquisition Unit 202 Engine control unit 203 Bitmap transmission unit 301 Print processing unit 400 Reading device 401 Read image acquisition unit 402 Master image processing unit 403 Inspection control unit 404 Comparison inspection unit 410 Paper discharge tray 421 Low-value multi-value conversion processing unit 422 Resolution conversion processing unit 423 Color conversion processing unit 424 Image output processing unit 431 Information input unit 432 Difference image acquisition unit 433 Defect determination unit 434 Flat region determination unit 435 Banding determination unit 436 Controller communication unit

Japanese Patent Laid-Open No. 2005-43769

Claims (9)

An image inspection apparatus for inspecting a read image obtained by reading an image formed and output on a recording medium,
A read image acquisition unit that acquires a read image generated by reading an image formed and output; and
A density change acquisition unit that acquires a change in a value related to the density of the read image according to a position in a direction in which the recording medium is conveyed based on the acquired read image;
A concentration fluctuation processing unit for removing fluctuations of a predetermined frequency or more among fluctuations occurring in the change of the value related to the density;
An inflection for detecting an inflection point, which is a point at which the direction of change according to the position in the direction in which the recording medium is conveyed changes with respect to a change in the value related to the density from which fluctuations of the predetermined frequency or more have been removed. A point detector;
An image including a periodic defect determination unit that determines, based on the detected inflection point, that a defect that periodically occurs in the transport direction of the recording medium occurs in the read image. Inspection device. The said periodic defect determination part determines that the defect which generate | occur | produces periodically occurs based on the variation | change_quantity of the value regarding the said density | concentration between the said inflection points. Image inspection equipment. The inflection point detection unit has a difference in value related to the density between adjacent points among the points where the direction of change according to the position in the direction in which the recording medium is conveyed is equal to or greater than a predetermined threshold. The image inspection apparatus according to claim 1, wherein a certain point is detected as the inflection point. The image according to any one of claims 1 to 3, wherein the density fluctuation processing unit determines the frequency at which the fluctuation is removed according to the size of the read image to be inspected. Inspection device. The density variation processing unit is a distance determined according to a size of the read image to be inspected, and based on an observation distance that is a distance from a viewpoint when a human views the read image to the read image. The image inspection apparatus according to any one of claims 1 to 4, wherein the frequency at which fluctuation is removed is determined. A flat area determination unit that detects a flat area in which a change in shading for each pixel is gentle in an image to be output.
The density change acquisition unit acquires a change in a value related to the density of the read image according to a position in a direction in which the recording medium is conveyed based on an area detected as the flat area in the read image. The image inspection apparatus according to any one of claims 1 to 5. A differential image acquisition unit that acquires a differential image configured by a difference value between a pixel that constitutes an inspection image to be compared with the read image and a pixel that constitutes the read image for the inspection of the read image; ,
The density change acquisition unit acquires a change in a value related to the density of the difference image according to a position in a direction in which the recording medium is conveyed based on the difference image. The image inspection apparatus according to claim 1. An image inspection system for inspecting a read image obtained by reading an image formed and output on a recording medium,
An image forming unit that performs image forming output on the recording medium;
An image reading unit that reads an image formed on the recording medium and generates a read image;
A read image acquisition unit for acquiring the generated read image;
A density change acquisition unit that acquires a change in a value related to the density of the read image according to a position in a direction in which the recording medium is conveyed based on the acquired read image;
A concentration fluctuation processing unit for removing fluctuations of a predetermined frequency or more among periodic fluctuations occurring in the change of the value related to the density;
An inflection for detecting an inflection point, which is a point at which the direction of change according to the position in the direction in which the recording medium is conveyed changes with respect to a change in the value related to the density from which fluctuations of the predetermined frequency or more have been removed. A point detector;
An image including a periodic defect determination unit that determines, based on the detected inflection point, that a defect that periodically occurs in the transport direction of the recording medium occurs in the read image. Inspection system. An image inspection method for inspecting a read image obtained by reading an image formed and output on a recording medium,
Acquire a read image generated by reading the image output
Based on the obtained read image, obtain a change in value relating to the density of the read image according to the position in the direction in which the recording medium is conveyed,
Removing fluctuations of a predetermined frequency or more from periodic fluctuations occurring in the change of the value related to the concentration,
Detecting an inflection point, which is a point at which the direction of change according to the position in the direction in which the recording medium is conveyed changes with respect to a change in the value related to the density from which fluctuations of the predetermined frequency or more have been removed,
An image inspection method comprising: determining, based on the detected inflection point, that a defect that occurs periodically in the transport direction of the recording medium has occurred in the read image.