Anomaly detection device, anomaly detection method, and program

Abstract

The present invention provides an anomaly detection device, an anomaly detection method, and a program that enable easier and more appropriate evaluation of reading anomalies. [Solution] The abnormality inspection device comprises a first setting unit, a second setting unit, an evaluation unit, and an output unit. The first setting unit detects a streak-like first abnormality caused by the image reading unit from a first read image by the image reading unit and sets a first streak intensity for the first abnormality. The second setting unit detects an abnormality from a second read image of the recording medium by the image reading unit and sets a second streak intensity for the detected streak-like second abnormality. The evaluation unit evaluates whether a streak-like third abnormality has occurred on the recording medium at a location overlapping with the first abnormality, based on the first and second streak intensities. The output unit outputs an evaluation result report.

Description

Technical Field

[0001] This invention relates to an abnormality inspection apparatus, an abnormality inspection method, and a program.

Background Art

[0002] Conventionally, there is a technique of detecting a formation defect of an image by reading an image formed by an image forming apparatus (printer) with an image reading apparatus (scanner). On the other hand, when there is dirt or an abnormality in parts of the image reading apparatus, streak-like abnormalities (scan streaks) occur depending on the recording medium that is scanned relatively to the dirt. When the formation defect of the image overlaps with the scan streak, it causes the formation defect to be overlooked. Therefore, a technique for separately identifying scanner streaks is known (Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, the conventional technique is too laborious for real-time processing such as forming an image and cannot perform identification processing easily.

[0005] An object of this invention is to provide an abnormality inspection apparatus, an abnormality inspection method, and a program that can more easily and appropriately evaluate a reading abnormality.

Means for Solving the Problems

[0006] To achieve the above object, one aspect of the present invention is a first setting unit that detects a streak-like first abnormality caused by the image reading unit from a first read image by the image reading unit and sets a first streak intensity for the first abnormality; A second setting unit detects an anomaly from the second read image of the recording medium by the image reading unit and sets a second streak intensity for the detected streak-like second anomaly, An evaluation unit that evaluates whether a third abnormality in the form of streaks has occurred at a location on the recording medium that overlaps with the first abnormality, based on the first and second streaks; An output unit that outputs the results report of the evaluation, It is an abnormality detection device equipped with [specific features / features].

[0007] Another aspect of the present invention is: A first setting step involves detecting a streak-like first anomaly caused by the image reading unit from a first read image obtained by the image reading unit, and setting a first streak intensity for the first anomaly. A second setting step involves detecting an anomaly from the second read image of the recording medium by the image reading unit, and setting a second streak intensity for the detected streak-like second anomaly. An evaluation step to evaluate whether a third abnormality in the form of streaks has occurred at a location on the recording medium that overlaps with the first abnormality, based on the first and second streaks. Output step to output the results report of the evaluation, This is an abnormality detection method that includes [specific methods].

[0008] Another aspect of the present invention is: Computers A first setting means for detecting a streak-like first anomaly caused by the image reading unit from a first read image by the image reading unit, and for setting a first streak intensity for the first anomaly, A second setting means detects an anomaly from the second read image of the recording medium by the image reading unit and sets a second streak intensity for the detected streak-like second anomaly. An evaluation means for evaluating whether a third abnormality in the form of streaks has occurred at a location on the recording medium that overlaps with the first abnormality, based on the first and second streaks. Output means for outputting the results report of the evaluation, This is a program that makes it function as such. [Effects of the Invention]

[0009] The present invention provides an abnormality inspection device, an abnormality inspection method, and a program that can more easily and appropriately evaluate reading abnormalities. [Brief explanation of the drawing]

[0010] [Figure 1] This is a front view showing an image forming system including an image inspection device. [Figure 2] This is a block diagram showing the functional configuration of an image inspection device. [Figure 3] This figure shows an example of streak detection. [Figure 4] This is a flowchart showing the control procedure for image inspection control processing. [Figure 5] This figure provides a schematic overview of an example of a detection report for streak abnormalities. [Figure 6] This figure provides a schematic overview of an example of a detection report for streak abnormalities. [Modes for carrying out the invention]

[0011] Hereinafter, embodiments of the present invention will be described based on the drawings. Figure 1 is a front view showing an image forming system 1 including the image inspection apparatus 30 of this embodiment.

[0012] The image forming system 1 includes an image forming apparatus 10, a relay device 20, an image inspection device 30, an ejection device 40, and a post-processing device 50, etc. The transport path of the recording medium from the image forming apparatus 10 to the post-processing device 50 is shown by a dashed line. For illustrative purposes, gaps are shown between each device, but the devices may be connected without any gaps.

[0013] The image forming apparatus 10 sequentially supplies a recording medium, such as paper, on which an image is to be formed, and forms an image on the surface of the supplied recording medium by an electrophotographic method. The image data to be formed may be acquired from an external device via a communication line. The image forming apparatus 10 may be obtained by applying toner to a position on the recording medium determined based on the acquired image data and thermally fixing the applied toner. The image forming apparatus 10 can form a color image, for example, using four colors of toner: C (cyan), M (magenta), Y (yellow), and K (black). The application of toner to the recording medium may be performed by a tandem method using separate intermediate transfer belts. Alternatively, the toner may be applied to the recording medium by other methods. In such an image forming apparatus 10, usually, a latent image is formed while scanning in the width direction of the recording medium perpendicular to the conveyance direction of the recording medium. That is, the width direction is parallel to the scanning direction. Accordingly, the conveyance direction is also called the sub-scanning direction. The image forming apparatus 10 has a tray for storing the recording medium and the like. There may be a plurality of trays according to different sizes of the recording medium and the like. The recording medium on which the image is formed is output to the relay device 20.

[0014] The image forming apparatus 10 may include an optical scanner and have a reading function. That is, the image forming apparatus 10 may be able to read the surface of a medium, such as paper, placed on a reading surface, such as a contact glass, by an optical scanner and generate image data. Further, the image forming apparatus 10 may have a copying function of forming an image on the recording medium as described above based on the generated image data.

[0015] The image forming apparatus 10 includes a display unit 11 and an operation reception unit 12. The display unit 11 has a digital display screen and can display menus, statuses, etc. related to image formation. As will be described later, a report indicating the inspection result of image abnormality can be displayed on the display unit 11. The display unit 11 may further have an LED lamp or the like. The LED lamp can be switched between a lit state, a blinking state, and an off state, or the lit color can be changed according to a predetermined status of the image forming system 1. The operation reception unit 12 receives input operations by a user or the like and outputs an operation signal indicating the received operation content. The operation reception unit 12 may have a touch panel that is positioned to overlap the display screen of the display unit 11. The control unit of the image forming apparatus 10 identifies the operation content based on the operation signal and the display content on the digital display screen when the operation signal is received, and performs processing according to the operation content. The operation reception unit 12 may further have switching elements such as push button switches, rotary switches, slide switches, and toggle switches. The switching elements may be used, for example, for on / off operations of functions such as turning on / off the main power supply and restricting states such as power saving states and standby modes. The operation reception unit 12 may have a keyboard such as a numeric keypad.

[0016] The relay device 20 relays and conveys the paper output from the image forming apparatus 10 to the image inspection apparatus 30, which is a subsequent device. The relay device 20 has a function of synchronizing with the conveyance speed of the recording medium conveyed from the image forming apparatus 10.

[0017] The image inspection device 30 is an abnormality inspection device of this embodiment, and inspects for abnormalities in the image by reading the image-forming surface of the recording medium after image formation. The image inspection device 30 has an imaging unit 31 as an image reading unit that captures the image-forming surface. The imaging unit 31 may have, for example, a line sensor 311 and a background plate 312. The line sensor 311 has multiple sensors, such as CCD sensors, arranged one-dimensionally in the width direction (second direction) perpendicular to the transport direction (first direction) of the recording medium. The line sensor 311 may perform detection in the wavelength bands of each RGB color. The background plate 312 is also used as a shading plate when calibrating the brightness values ​​of white and black backgrounds. Therefore, the background plate 312 can be selectively switched and arranged within the shooting range to have a white surface and a black surface. The width of the background plate 312, that is, the length in the direction perpendicular to the transport path of the recording medium, is set to be larger than the maximum size of the recording medium that can be image-formed by the image forming device 10.

[0018] The image inspection device 30 has rollers that move the recording medium in a direction along the transport path outside the imaging range. The image inspection device 30 can obtain two-dimensional image data by sequentially photographing the recording medium as it is moved relative to the line sensor 311 in the transport direction by the rotational movement of the rollers. If the image forming apparatus 10 is capable of forming images on both sides of the recording medium, the imaging unit 31 may have a first line sensor that images one side of the recording medium and a second line sensor that images the opposite side. The image inspection device 30 may further have a colorimeter.

[0019] The image inspection device 30 outputs the obtained two-dimensional image data as inspection image data to the controller 32 (see Figure 2). The controller 32 is a hardware processor that analyzes the inspection image and detects image abnormalities. The detection of abnormalities will be described later. The controller 32 may be a microcontroller with a dedicated ASIC (Application Specific Integrated Circuit), or it may have a general-purpose CPU. The operation of the controller 32 can be set and started according to the control of the image forming apparatus 10. A result report (report) showing the abnormality detection result may be output to the image forming apparatus 10 and displayed on the display unit 11, for example. The recording medium after imaging is output to the ejection device 40.

[0020] The discharge device 40 may separate specific recording media used for inspection from the normal transport path and discharge them separately into tray T1. Alternatively, the discharge device 40 may acquire image anomaly detection results from the image inspection device 30 and separate recording media in which anomalies have been detected from the normal transport path and discharge them into tray T1. Recording media traveling along the normal transport path are output to the post-processing device 50.

[0021] The post-processing device 50 performs various post-processing operations on the image-formed recording medium output from the discharge device 40 and discharges it to the tray T2. Post-processing may include, in part or in whole, sorting, stapling, punching, folding, and binding.

[0022] Figure 2 is a block diagram showing the functional configuration of the image inspection device 30. The image inspection device 30 includes, in addition to the imaging unit 31 and controller 32 described above, a storage unit 33, a communication unit 34, a transport unit 35, and the like. The imaging unit 31 also has a background plate setting unit 313 that switches the background plate 312 between a white side and a black side. The white side and the black side may be sides of different plates, or opposite sides of a single plate. Alternatively, the white side and the black side may be located on different parts of one surface of a plate.

[0023] The memory unit 33 is a non-volatile memory that stores the program 331 and setting data related to image inspection, as well as the inspection report. The non-volatile memory may be flash memory or an HDD (Hard Disk Drive). The setting data includes rank-corresponding data 332 and tolerance standard data 333. The rank-corresponding data 332 and tolerance standard data 333 will be described later.

[0024] The communication unit 34 controls communication with the image forming apparatus 10 and the ejection device 40. Inspection requests and other information are received from the image forming apparatus 10 via the communication unit 34. Inspection result reports are transmitted to the image forming apparatus 10 via the communication unit 34. An ejection command for a recording medium in which an abnormality has been detected is transmitted to the ejection device 40 via the communication unit 34. The communication standard used by the communication unit 34 only needs to be capable of bidirectional communication of the above transmitted and received content with the image forming apparatus 10 at the necessary data rate.

[0025] The transport unit 35 has the rollers and their drive units described above. The transport unit 35 moves the recording medium received from the relay device 20 so that it passes through the imaging range of the imaging unit 31, and then sends it to the discharge device 40. The transport speed of the discharge device 40 may be adjustable according to the transport speed during image formation in the image forming apparatus 10.

[0026] Next, the image inspection of this embodiment will be described. The image inspection device 30 inspects whether the colorant has been properly applied and whether an image has been formed in the image forming apparatus 10. Image abnormalities can take on specific shapes such as streaks, lumps, or spots, depending on the cause. Abnormalities can also include abnormalities where toner is not applied, abnormalities where toner or dirt adheres, and abnormalities where toner application is uneven or inconsistent. By photographing the surface of the recording medium after image formation and detecting the shape and color of these abnormalities from the captured data, the location and cause of the abnormality can be identified.

[0027] On the other hand, even if there are no abnormalities in the formed image, abnormalities may appear in the captured data due to shooting abnormalities of the image inspection device 30 or the reflection of dirt. If there is an abnormality in the imaging unit 31 that prevents it from moving relative to the line sensor 311, a streak-like abnormality will be detected continuously at a certain position in the width direction as the recording medium moves in the transport direction. This streak-like abnormality that appears apparent due to the imaging unit 31 is the first abnormality of this disclosure and will be referred to as a scan streak. Abnormalities that prevent movement may include dirt on the light incident surface of the line sensor 311, for example, on the dustproof cover glass. Note that dirt on the background plate 312 is invisible behind the recording medium M when the recording medium M is being photographed, so it is only photographed when the recording medium M is not present.

[0028] These scan lines may overlap with FD lines (Feed Direction), which are streak-like abnormalities in the image that extend in the direction along the transport path. As a result, FD lines may be hidden by scan lines and overlooked. The image inspection device 30 evaluates the possibility that a third type of abnormality, an FD line, has occurred on the recording medium in the captured image of the recording medium at a location that overlaps with the scan lines.

[0029] Figure 3 shows an example of streak detection. The vertical direction in the diagram represents the transport direction of the recording medium. In the example in Figure 3(a), a black surface is shown as the background plate 312. In the example in Figure 3(b), a white surface is shown as the background plate 312. Note that the black and white surfaces do not actually need to extend over the length of the recording medium M in the transport direction. It is sufficient that the same area of ​​the background plate 312 is continuously reflected behind the transported recording medium M, as long as the background plate 312 is stationary within the shooting range of the line sensor 311. White streaks Sw and black streaks Sk appear parallel to the transport direction across the entire recording medium M.

[0030] Meanwhile, test images are positioned on the recording medium M, such as shading images for correcting density unevenness of each color captured by the imaging unit 31, and adjustment images for adjusting density unevenness of the formed image by the image forming apparatus 10. These test images may be halftone images of predetermined density gradations. For example, a halftone image Iy of color Y, a halftone image Im of color M, a halftone image Ic of color C, and a halftone image Ik of color K are arranged in order in the transport direction. Of these, a streak Sy extending along the transport direction appears in the range of halftone image Iy. A streak Sc extending along the transport direction appears in the range of halftone image Ic. White streaks Sw and black streaks Sk extending outside of halftone images Iy, Im, Ic, and Ik are scan streaks, and streaks Sy and streaks Sc extending only within the halftone image of any of the colors are FD streaks. An anomaly, including FD streaks (second anomaly), is detected from the second read image obtained by capturing the recording medium including the test images with the line sensor 311.

[0031] Scan streaks may be identified from a first read image obtained by photographing the background plate 312 without a recording medium M. In this case, in order to distinguish them from dirt on the background plate 312, the background plate 312 may be moved relative to the shooting range of the line sensor 311 in the transport direction during shooting. White streaks Sw are identified from the captured image when the black side (black background) of the background plate 312 is displayed. Black streaks Sk are identified from the captured image when the white side (white background) of the background plate 312 is displayed.

[0032] FD streaks, specifically streaks Sy and Sc, are identified based on the distribution of density distributions in the width direction of each halftone image Iy, Im, Ic, and Ik, e.g., relative variability.

[0033] FD streaks and scan streaks have different causes, and therefore require completely different treatments. FD streaks degrade the quality of the output image, so they need to be addressed. On the other hand, scan streaks often produce a normal output image even without any treatment. When scan streaks and FD streaks overlap at the same position in the width direction, FD streaks may be overlooked and not treated, resulting in an uncorrected image. In Figure 3(b), streaks Sy overlap with scan streaks Sw, making them difficult to identify.

[0034] In this embodiment, the image forming apparatus 10 forms a test image in which halftone images of each CMYK color shown in Figure 3 are arranged in a strip. The image inspection apparatus 30 captures this test image and performs inspection. As described above, the black and white background plates 312 may be photographed separately beforehand without the recording medium M. Alternatively, the portion of the background plate 312 outside the range of the recording medium M that is exposed may also be photographed during transport. The image inspection apparatus 30 identifies scan lines based on the distribution of luminance gradation in the width direction of the captured image of the background plate 312. That is, the luminance gradation shows a narrow peak-like change at the point of the line. The luminance gradation may be acquired along a line in the width direction. Alternatively, the average luminance value over a certain width range in the transport direction may be obtained at each position in the width direction.

[0035] Furthermore, if the image inspection device 30 is equipped with a colorimeter, the relationship between the colorimeter's measured values ​​and the luminance values ​​obtained from the imaging unit 31 may be determined in advance. This allows for appropriate quantitative evaluation of the variation in density gradation based on the luminance values.

[0036] The image inspection device 30 sets a rank (first streak intensity) for each identified scan streak according to its size. The size includes density (brightness) and width. At this time, the white streak rank, which is the rank for white streaks, is identified from the image captured on the black background plate 312. The black streak rank, which is the rank for black streaks, is identified from the image captured on the white background plate 312. Depending on the color of the scan streak, it may be identified as either a white streak or a black streak. The white streak rank and the black streak rank are determined independently. Therefore, different streaks with the same white streak rank may be assigned different black streak ranks. Also, different streaks with the same black streak rank may be assigned different white streak ranks. Although not particularly limited, the rank may be set so that whiter streaks have a larger positive value and blacker streaks have a larger negative value in absolute value, according to the brightness gradation.

[0037] The image inspection device 30 analyzes each halftone image to detect abnormalities. The image inspection device 30 extracts streak-like abnormalities from among the abnormalities. For each identified streak, the image inspection device 30 sets a streak rank as a second streak intensity separately for each color. The streak rank is determined based on the width of the streak and the relative change in brightness value. In addition, if FD streaks shorter than the length in the transport direction of the halftone image are considered, the length of those FD streaks may also be considered in the streak rank. This identified streak rank is set regardless of whether it is a scan streak or an FD streak. The streak rank for a scan streak is independent of the rank of the scan streak. There is a predetermined correlation between the streak rank for the same scan streak and the rank of the scan streak. This correlation (information) is stored in the storage unit 33 in advance as rank correspondence data 332. That is, based on the rank correspondence data 332, the streak rank in each halftone image corresponding to the rank of the scan streak is obtained as a reference value.

[0038] When scan streaks and FD streaks overlap at the same position in the width direction, a deviation occurs in the halftone image where the overlap occurs, relative to the reference value of the streak rank obtained by the correlation between the rank of the scan streaks and the above-mentioned correlation. Therefore, if this deviation is larger than the reference value, it is presumed that FD streaks are overlapping with scan streaks. That is, the image inspection device 30 evaluates whether FD streaks, as a third abnormality, have occurred on the recording medium, overlapping with scan streaks, based on the rank of the scan streaks and the rank of the FD streaks. If it is presumed that FD streaks have occurred, the streak rank attributable to the FD streaks is obtained as the third streak intensity based on the difference between the above-mentioned streak rank and the reference value.

[0039] The evaluation results of scan streaks and FD streaks obtained in this manner are further judged to be within an acceptable range for the user of the formed image. The predetermined criteria for the acceptance judgment may be set separately for each CMYK color of FD streaks and scan streaks. The acceptance judgment criteria can be set for each user of the output image. The user of the output image may be the same as the owner of the image forming system 1. Alternatively, the image forming system 1 may be used by multiple people, and each may set their own acceptance judgment criteria. The settings may be variable. Furthermore, acceptance or rejection may be determined according to the target image to be formed. The visibility of streaks and color unevenness differs depending on the type of image, such as whether the target image is text, a figure or sign, or an image. The image inspection device 30 acquires the image type as data for these target images and may change the settings according to the acquired data. Furthermore, the image inspection device 30 may divide the visibility of FD streaks within the image into stages in more detail and dynamically determine the acceptance judgment criteria according to the position within the image. Furthermore, criteria and judgments may be set and judged not only based on the target image, but also by contextual analysis including printing conditions and operating parameters. In addition, the criteria for acceptable judgment may be determined using a machine learning model. The detected image of FD streaks and / or scan streaks is used as the input image, and training data is generated with the acceptableness of each streak in the input image as the ground truth data. By inputting this training data into a machine learning model and training it, a trained model is obtained that outputs acceptableness for the input image. Note that the input image may or may not include information on streak position and / or streak rank. If information on streak position and streak rank is included, information on the overlap between FD streaks and scan streaks may also be included.

[0040] The criteria for determining acceptance may be stored in the storage unit 33 in advance as acceptance criterion data 333, with subjective expressions such as "strict," "normal," and "lenient" associated with specific standard values. In this case, the user or the administrator setting up the image inspection device 30 for the user may determine the criteria using the subjective expressions as requested by the user. The standard value does not have to be just one to determine acceptance or rejection. For example, based on two standard values, the judgment may be made in three stages: acceptable (OK), difficult to accept (almost NG), and unacceptable (NG). Note that such changes in settings may be applied to FD streaks, while fixed criteria may be set for scan streaks.

[0041] The result of the judgment is output as a result report. The output result report may be sent to the image forming apparatus 10 and displayed by the display unit 11, as described above. The report may be formed according to a predetermined format. In this case, if it is estimated that FD lines overlap with scan lines, these scan lines and the estimated FD lines may be shown in correspondence. The correspondence may be, for example, simply displayed side by side, connected by arrows, indicated by matching colors, or noted as an annotation that they are in the same location in the width direction. In this way, the estimated overlapping portion may be displayed separately from the specific location of the simple FD lines or scan lines.

[0042] The report may include captured images. In the captured images, detected areas such as FD streaks may be indicated by markers such as arrows. The markers may have different colors or thicknesses for FD streaks and scan streaks. In addition, the rank of the scan streaks or streak rank may be displayed in accordance with the markers. The streak rank displayed here is the streak rank (second streak intensity) determined for the FD streaks if the FD streaks do not overlap with the scan streaks. If the FD streaks overlap with the scan streaks, it is the third streak intensity determined based on the deviation from the reference value from the streak rank. These are collectively referred to as the fourth streak intensity of the fourth abnormality. The process of identifying the fourth abnormality is the operation of the identification unit of the controller 32 in this embodiment. Alternatively, the results output to the report may include only information on streaks that were judged as NG by the acceptance judgment. The report may be generated in a size that can be displayed by the display unit of the image forming apparatus 10, or it may be a size that requires scrolling of the display range or switching pages. In addition to or separately from this, the report data may be output or image forming output in a document format such as PDF. Furthermore, the report output format may be customized by the user.

[0043] Figure 4 is a flowchart showing the control procedure for the image inspection control process performed in the image inspection apparatus 30 of this embodiment. This image inspection control process is executed when an image inspection request is received from the image forming apparatus 10.

[0044] The controller 32 sets the background plate 312 so that the black side can be photographed using the background plate setting unit 313 (S1). The controller 32 causes the line sensor 311 to photograph the background plate 312 and acquires the photographic data (S2). The controller 32 sets the background plate 312 so that the white side can be photographed using the background plate setting unit 313 (S3). The controller 32 causes the line sensor 311 to photograph the background plate 312 and acquires the photographic data (S4). The photographic data may be two-dimensional data of position in the width direction and photographic time, or one-dimensional data of position in the width direction integrated with respect to photographic time.

[0045] The controller 32 performs a predetermined analysis process on each of the two captured data images and detects scan streaks (S5). For example, the controller 32 extracts abnormal portions as scan streaks in the brightness value distribution in the width direction where there is a large deviation from the standard. White streaks are detected based on an increase in brightness value in the captured image of a black surface. Black streaks are detected based on a decrease in brightness value in the captured image of a white surface. The standard may be the average value of the brightness value across the entire width direction including the streaks. Alternatively, the standard may be perfect white (brightness value 255) or perfect black (brightness value 0). The threshold that defines a large deviation may be a fixed value, or it may be determined based on the standard deviation of the brightness value distribution. To confirm that it is not due to the influence of temporary noise, the controller 32 may check whether a continuous change in brightness value occurs over the shooting time. Alternatively, the threshold or shooting time may be set so that the change in the average brightness value does not reach the threshold for noise below a certain time. Based on the detection results, the controller 32 identifies the scan streaks (S6). The processing in step S6 is the operation of the controller 32 as the first setting unit in this embodiment. Furthermore, the processing in step S6 corresponds to the first setting step in the abnormality inspection method of this embodiment, and corresponds to the first setting means that the program 331 of this embodiment causes the computer to execute.

[0046] The controller 32 identifies the width of each extracted scan streak and its brightness value or the amount of change in brightness value. Based on these identified values, the controller 32 determines the rank of the scan streak (S7). If no scan streak is identified in step S6, the controller 32 omits the process in step S7.

[0047] The controller 32 waits for the recording medium M to be sent from the image forming apparatus 10 to the image inspection apparatus 30 via the relay device 20. As described above, CMYK four-color halftone images are formed sequentially in the transport direction on the recording medium M as test images. The controller 32 causes the imaging unit 31 to photograph the surface of the obtained recording medium and acquires the photographic data (S8).

[0048] The controller 32 performs a predetermined image analysis process (S9). The image analysis process includes at least a process for recognizing the range of the recording medium. Information such as the formation position of each halftone image within the recording medium may be set and held in advance. Alternatively, the process for recognizing the range of the halftone image may be included in the image analysis process.

[0049] The controller 32 detects non-uniformity in the luminance value distribution within a range that includes each halftone image of the test image in the transport direction. In this case, the luminance values ​​may be obtained in each RGB wavelength band and only the data in the wavelength bands suitable for each color of the halftone image, i.e., CMYK, may be used. The controller 32 extracts portions in the luminance value distribution that deviate significantly from the standard and extracts them as image anomalies. That is, image anomalies are identified for each CMYK color. In particular, the controller 32 detects portions where the image anomaly continues for a standard length or longer in the transport direction as FD streaks (S10). The processing in step S10 corresponds to the operation of the controller 32 as the second setting unit in this embodiment. Furthermore, the processing in step S10 corresponds to the second setting step in the anomaly inspection method of this embodiment and corresponds to the second setting means that the program 331 causes the computer to execute.

[0050] The controller 32 identifies the width of the FD streak and its brightness value or the amount of change in brightness value. Furthermore, if the controller 32 considers partial FD streaks shorter than the length of the halftone image in the transport direction, it identifies the length of such FD streaks. Based on these identified values, the controller 32 determines the streak rank (S11).

[0051] The controller 32 determines whether or not the scan streak has been identified (S12). If it is determined that the scan streak has not been identified (S12;N), the controller 32 proceeds to step S14.

[0052] If it is determined that a scan streak has been identified (S12;Y), the controller 32 identifies the streak rank at the widthwise position corresponding to the location of the scan streak in each halftone image. The controller 32 identifies this streak rank as the rank of the scan streak and determines whether the deviation from the rank of the scan streak is greater than or equal to a standard. Based on the determination result regarding the deviation, the controller 32 calculates the possibility that an FD streak is superimposed on the scan streak (S13). The processing in step S13 corresponds to the operation of the controller 32 as an evaluation unit in this embodiment. Furthermore, the processing in step S13 corresponds to the evaluation step in the abnormality inspection method of this embodiment, and the program 331 of this embodiment causes the computer to function as an evaluation means. If there is a possibility that an FD streak is superimposed, the controller 32 may estimate the streak rank of the superimposed FD streak. Then, the processing of the controller 32 proceeds to step S14.

[0053] When the process moves to step S14, the controller 32 makes a judgment on whether each identified FD streak is acceptable or not (S14). The controller 32 obtains a standard value for the rank according to the user from the acceptable standard data 333, and if the rank is worse than the standard value, it may determine that the streak of that rank is at an unacceptable level.

[0054] Based on the detection results of each streak and the results of the acceptance judgment described above, the controller 32 creates an inspection result report (report) to be provided to the user (S15). As described above, the report may include captured images. In this case, the controller 32 may indicate the detected streaks on the captured images using arrows or the like, and the detected rank and acceptance judgment results may be shown in correspondence with the arrows.

[0055] The controller 32 outputs the created report to an output destination corresponding to the output target via the communication unit 34 (S16). In the case of display data for a display, the data of the display content is output to the image forming apparatus 10. In the case of PDF format, the report may be output directly to the storage unit 33 and stored therein. In addition to or instead of this, the report may be output directly via the communication unit 34 or further via the image forming apparatus 10 to an external specific person or management server. When the report is output externally, the report may be encrypted. The processing in step S16 is the operation of the controller 32 as an output unit in this embodiment, and corresponds to the output step and output means of this embodiment. Then, the controller 32 terminates the image inspection control processing.

[0056] Figures 5 and 6 are schematic diagrams illustrating some examples of streak abnormality detection reports. As shown in Figure 5, for example, the overall results may be summarized on the display unit 11 at the beginning of the image inspection report.

[0057] As shown in Figure 6, in captured images where an anomaly is detected, markers such as arrows and ranks may be added to indicate the location of the anomaly, making it possible to identify the specific location of the anomaly. For example, a downward arrow may be shown at the location where a streak extending in the transport direction, i.e., the vertical direction in the figure, is detected. Scan streaks are shown in the margins of the test image. Here, arrow Aw, representing a white streak, and arrow Ak, representing a black streak, are shown near the top. Next to the arrows, the type of streak and the determined rank are displayed. FD streaks are shown within each halftone image. In halftone image Iy, arrow Ay, indicating a yellow streak, is shown along with its streak rank. In halftone image Im, arrow Am, indicating a magenta streak, is shown along with its streak rank. In halftone image Ic, arrows Ac1 and Ac2, indicating cyan streaks, are shown along with their streak ranks.

[0058] Here, there are two types of arrows. For example, a hollow arrow could represent a "almost unacceptable" level, while other arrows could represent an "unacceptable" level.

[0059] The FD streak indicated by arrow Ac2 overlaps with the scan streak indicated by arrow Aw. The above processing estimates the overlap between the scan streak and the FD streak. Specific details of such inspection results may be included in the report as text.

[0060] FD streaks are an abnormality of the image forming apparatus 10 and can be improved by adjusting the image data. On the other hand, scan streaks are not an abnormality caused by the image, so measures such as cleaning the image inspection apparatus 30 are required. In particular, if the overlap of scan streaks and FD streaks is suspected, a warning notification of the need for cleaning may be issued in order to more accurately determine whether the FD streaks are acceptable or not. These countermeasures may be written in the margins of area Rw. This detection result may be included in the output report. Alternatively, the detection result may be included in the output of a document such as a PDF. When the report is displayed on the display unit 11, the detection result may not be included in the report, or it may be output in an adjustable manner so that it can be easily viewed, such as by dividing and displaying enlarged views of each part.

[0061] As a result, the controller 32 terminates the image inspection control process.

[0062] As described above, the image inspection device 30, which is one aspect of the abnormality inspection device of this embodiment, includes a controller 32. The controller 32, as a first setting unit, detects scan streaks originating from the image inspection device 30 from the first read image by the image inspection device 30 and sets a rank for the scan streaks. The controller 32, as a second setting unit, detects abnormalities from the second read image of the recording medium M by the imaging unit 31 and sets a streak rank for the detected streaky second abnormality. The controller 32, as an evaluation unit, evaluates whether FD streaks have occurred at a location on the recording medium M that overlaps with the scan streaks, based on the rank of the scan streaks and the streak rank. The controller 32, as an output unit, outputs an evaluation result report. With such an image inspection device 30, it is possible to quantitatively evaluate scan streaks and streaks detected from the image and evaluate the possibility that FD streaks overlap with scan streaks. Therefore, the image inspection device 30 can reduce the chances of overlooking FD streaks and evaluate reading abnormalities more easily and appropriately.

[0063] Furthermore, a test image with a predetermined density gradation may be formed on the recording medium M by the image forming apparatus 10. The second read image may be obtained while moving the recording medium M relative to the imaging unit 31 in the transport direction. The controller 32, acting as a second setting unit, may detect a second abnormality based on the distribution of brightness values ​​in the width direction perpendicular to the transport direction in the test image. In this configuration, where the image is read by the line sensor 311 while the recording medium M is moved, scan streaks that appear to extend in the transport direction may occur due to problems on the line sensor 311 side. The image inspection apparatus 30 can more appropriately separate FD streaks from such scan streaks and reduce reading abnormalities.

[0064] Furthermore, the controller 32 may, as an evaluation unit, hold information on a reference value of the streak rank corresponding to the rank of the scan streak. The controller 32 may also evaluate the possibility that the FD streak is occurring in overlap with the scan streak based on the deviation from the reference value of the streak rank set as a second setting unit. By considering the possibility of the FD streak overlapping with the scan streak quantitatively in this way, the image inspection device 30 can further reduce the chances of overlooking FD streaks.

[0065] Furthermore, the controller 32, as an evaluation unit, may obtain the streak rank of the FD streaks themselves based on the difference between the streak rank and the reference value for FD streaks that overlap with scan streaks. In this way, even for FD streaks whose rank cannot be directly determined, the image inspection device 30 can estimate the degree to which they are abnormal to some extent. Therefore, the user can respond more appropriately.

[0066] Furthermore, the rank of a scan streak may be determined based on the width and brightness of the scan streak. The streak rank of a streak in a read image of a recording medium may also be determined based on the width and brightness of the streak. By determining the rank of a streak using parameters that are important for a streak, the image inspection device 30 can accurately calculate the degree of a streak with simple processing.

[0067] Furthermore, the streak rank of streaks in the image read from the recording medium may be determined by taking into account the length of the streaks. Since FD streaks do not necessarily continue uniformly in the transport direction, the length of FD streaks that are interrupted may be taken into consideration when determining the streak rank.

[0068] Furthermore, scan streaks may include black streaks appearing on a white background and white streaks appearing on a black background. The controller 32 may, as a first setting unit, separately assign ranks to black streaks and white streaks. The color of the streaks may also differ depending on the cause and location. Therefore, by handling bright white streaks and dark black streaks separately, the likelihood of overlooking scan streaks or overestimating / underestimating their impact is reduced.

[0069] Furthermore, the controller 32 may, as a specific unit, identify streak abnormalities on the recording medium M that do not overlap with scan streaks, and streak abnormalities identified as overlapping with scan streaks, as FD streaks on the recording medium M. In the results report, it may be determined whether or not the FD streaks are acceptable based on predetermined criteria. That is, the image inspection device 30 may determine, based on the streak rank or the like, whether the FD streaks identified as described above cause an unacceptable decrease in image quality for the target image to be output. If the streaks are minute and do not affect the required image quality, they may be left as is, but if the output can only meet the required image quality, then measures to fill in the streaks are necessary. By reporting such a judgment along with the results, the image inspection device 30 can provide the user with not only the detection results but also suggestions on whether or not further action is necessary. Therefore, the user can take the necessary action more easily and quickly.

[0070] Furthermore, the predetermined criteria for the above judgment may be variable. The required level of image quality varies depending on the user. Moreover, even for the same user, the required level of image quality may change depending on the target image. Therefore, by making the judgment criteria variable, it is possible to provide more accurate judgment results.

[0071] Furthermore, the controller 32, acting as an evaluation unit, may acquire data of the target image to be formed and determine predetermined criteria according to that data. In other words, the criteria for judgment may be determined directly from the data of the target image. The required level of image quality differs between text images and landscape photographs. Also, even for target images of the same type, the visibility of streaks differs depending on the color tone. Therefore, by directly determining criteria using the target image, the image inspection device 30 can output more accurate judgment results.

[0072] Furthermore, the controller 32, as an evaluation unit, may dynamically determine predetermined criteria based on the data of the target image, according to the position within the read image during inspection. That is, if the target image contains a mixture of text and image portions, or if the image contains a mixture of light-colored and dark-colored images, predetermined criteria may be set for each part of the image. This allows the image inspection device 30 to more accurately determine the impact of FD streaks on the image quality of the target image.

[0073] Furthermore, in the results report, if the scan streaks and FD streaks overlap in position on the scanned image, they may be shown in correspondence. This allows the user to understand that the FD streaks are not visible independently. Therefore, the user will be motivated to remove the cause of the scan streaks through cleaning, etc.

[0074] Furthermore, the results report distinguishes between scan streaks and FD streaks that are correlated with each other, and FD streaks on the read image that do not overlap with scan streaks, as well as scan streaks that do not overlap with FD streaks. By showing both scan streaks and FD streaks separately when they overlap, it becomes clearer that the presence of FD streaks is an estimation, and this provides motivation for cleaning the cause of scan streaks.

[0075] Furthermore, the results report may include images of test images. These test images may show markers indicating the locations where scan streaks and FD streaks were detected. By explicitly indicating the location of the detected streaks, users can easily visualize the streaks without having to search for them.

[0076] Furthermore, the controller 32 may, as a specific unit, treat the streak rank of FD streaks that do not overlap with scan streaks and the estimated streak rank of FD streaks that overlap with scan streaks as a single streak rank. The result report may include the display of the rank of the scan streaks and the streak rank of the FD streaks. This allows the image inspection device 30 to show the degree of streaks objectively, rather than relying solely on the user's subjective judgment.

[0077] Furthermore, the controller 32 may have a trained model as an evaluation unit that defines acceptable standard values ​​for the evaluation results. Since the judgment of whether something is good or bad ultimately depends on objective impressions, it is difficult to make a judgment using only simple parameters. By using a trained model for judgment, the image inspection device 30 can make a more situation-appropriate decision regarding the acceptance or rejection of streaks.

[0078] Furthermore, the predetermined format for outputting the results report may be customizable. By allowing customization according to the items that users want to emphasize and those that they do not need to worry about as much, the image inspection device 30 can output a report that is more suitable for the user.

[0079] Furthermore, the abnormality inspection method of this embodiment includes the following steps: (1) A first setting step in which scan streaks originating from the imaging unit 31 are detected from a first read image by the imaging unit 31, and a rank is set for the scan streaks. (2) A second setting step in which abnormalities are detected from a second read image of the recording medium M by the imaging unit 31, and a streak rank is set for the detected streaky abnormalities. (3) An evaluation step in which, based on the rank of the scan streaks and the streak rank, FD streaks are observed at the location on the recording medium M that overlaps with the scan streaks. (4) An output step in which an evaluation result report is output. According to the above anomaly inspection method, it is possible to quantitatively evaluate scan streaks and streaks detected from the image to assess the possibility that FD streaks overlap with scan streaks. Therefore, it is possible to reduce the chances of overlooking FD streaks and to evaluate reading anomalies more easily and appropriately.

[0080] Furthermore, by installing and running the program 331 related to the above-mentioned abnormality inspection method on a computer, it is possible to easily reduce the chances of overlooking FD streaks from the read image without requiring any special hardware.

[0081] It should be noted that the present invention is not limited to the embodiments described above, and various modifications are possible. For example, in the above example, the strength of the tendons was expressed as a tendon rank, but the name and calculation method may be arbitrary as long as quantitative evaluation is possible.

[0082] Furthermore, while the above describes detecting streaks using the capture data of test images, this is not the only method. Streaks may also be detected using the capture data of the target image. In this case, streaks will not be formed or detected in areas without images. However, since the image quality of areas without images in the target image does not deteriorate, streaks that do not affect image formation may be ignored in this case.

[0083] Furthermore, although the above only describes the detection of streak-like abnormalities, the image inspection device 30 may also perform the detection of abnormalities in images of other shapes in parallel.

[0084] Furthermore, while the streak rank was determined above based on the width and brightness of the streaks, it is not limited to this. Other factors may also be considered.

[0085] Furthermore, if it is estimated that scan streaks and FD streaks overlap, an index corresponding to the probability of the FD streaks existing in the first place may be calculated, in addition to or instead of directly determining the streak rank of the FD streaks. This index may be included in the output of the report.

[0086] Furthermore, the method of indicating lines in the report may be arbitrary. The direction of the arrows does not have to be in line with the lines. The type of arrow may differ from the example above. Also, arrows may be displayed in color.

[0087] Furthermore, although the above description assumes that the image forming apparatus 10 and the image inspection apparatus 30 are separate devices, this is not limited to this configuration. The image inspection apparatus may be included as part of the image forming apparatus 10. Also, the controller 32 of the image inspection apparatus 30 and the control unit of the image forming apparatus 10 may share a common hardware processor.

[0088] On the other hand, each of the above processes may be distributed and executed by the control units of multiple devices.

[0089] Furthermore, although the above describes how the recording medium M, which has been image-formed by the image forming apparatus 10, is automatically introduced into the image inspection apparatus 30 and an image is captured, the method is not limited to this. As described above, if the image forming apparatus has a scanner function, the image-formed recording medium M may be manually placed on the contact surface of the scanner by the user, and the image-formed surface of the recording medium M may be imaged. In this case, the background plate 312 may also be manually placed on the contact surface, and the white and black surfaces may be imaged, respectively.

[0090] Furthermore, although the above describes the recording medium M being transported and imaged while moving relative to a fixed line sensor 311, the method is not limited to this. The recording medium M may also be temporarily stationary, and the line sensor 311 may move relative to it in a direction parallel to the transport direction while image is taken.

[0091] On the other hand, the image inspection process may be performed independently of the image forming system 1 by an external device, which acquires the image data for the image forming surface of the recording medium M and the image data for each color surface of the background plate 312. The image forming system 1 may also acquire the judgment result regarding the acceptableness of streaks from the external device.

[0092] Furthermore, the report output may be uniform; in other words, it does not need to be customizable.

[0093] Furthermore, although the above explanation used the example of an image forming apparatus 10 forming an image using an electrophotographic method, it is not limited to this. The image forming apparatus 10 may also form images using other methods.

[0094] Furthermore, while the above description has used a storage unit 33 consisting of an HDD or non-volatile memory such as flash memory as an example of a computer-readable medium for storing the program 331 related to the control of image inspection of the present invention, the invention is not limited to these. Other computer-readable mediums that can be used include other non-volatile memories such as MRAM, and portable recording media such as CD-ROMs and DVD discs. In addition, a carrier wave can also be used as a medium for providing the program data of the present invention via a communication line. Furthermore, the specific configurations, processing operations, and procedures shown in the above embodiments can be modified as appropriate without departing from the spirit of the present invention. The scope of the present invention includes the scope of the invention described in the claims and its equivalents. [Explanation of symbols]

[0095] 1. Image forming system 10 Image forming apparatus 11 Display section 12 Operation reception section 20 Relay device 30. Imaging inspection equipment 31 Imaging Unit 311 Line Sensor 312 Background board 313 Background board setting section 32 controllers 33 Storage section 331 Programs 332 Rank-Compatible Data 333 Tolerance Criteria Data 34 Communications Department 35 Conveying section 40 Ejector 50 Post-processing equipment M recording medium Sc, Sy Suji Sk Black Stripes Sw White Stripes Sw scan lines T1, T2 trays

Claims

1. A first setting unit detects a streak-like first anomaly originating from the image reading unit from a first read image obtained by the image reading unit, and sets a first streak intensity for the first anomaly, A second setting unit detects an abnormality from the second read image of the recording medium by the image reading unit and sets a second streak intensity for the detected streak-like second abnormality, An evaluation unit that evaluates whether a third abnormality in the form of streaks has occurred at a location on the recording medium that overlaps with the first abnormality, based on the first and second streaks. An output unit that outputs the results report of the evaluation, An abnormality detection device equipped with the following features.

2. The recording medium has a test image with a predetermined density gradation formed on it by an image forming apparatus. The second read image is obtained while moving the recording medium in the first direction relative to the image reading unit. The second setting unit detects the second anomaly based on the distribution of brightness values ​​in the second direction perpendicular to the first direction in the test image. An abnormality inspection device according to claim 1.

3. The evaluation unit has information on a reference value for the second muscle strength corresponding to the first muscle strength, and evaluates the possibility that the third abnormality is occurring in conjunction with the first abnormality based on the deviation of the second muscle strength from the reference value set by the second setting unit. An abnormality inspection device according to claim 1.

4. The abnormality inspection device according to claim 3, wherein the evaluation unit obtains the third muscle strength of the third abnormality based on the deviation between the second muscle strength related to the third abnormality and the reference value.

5. The first streak intensity is determined based on the width and brightness of the first abnormality. The second streak intensity is determined based on the width and brightness of the second anomaly. An abnormality inspection device according to claim 1.

6. The abnormality inspection device according to claim 5, wherein the second streak strength is determined taking into consideration the length of the streak.

7. The first abnormality includes black streaks appearing on a white background and white streaks appearing on a black background. The first setting unit sets the first streak strength separately for the black streak and the white streak, respectively. An abnormality inspection device according to claim 1.

8. The system includes a identifying unit that identifies the second and third abnormalities on the recording medium at locations on the recording medium that do not overlap with the first abnormality, as a fourth abnormality on the recording medium. The abnormality inspection device according to claim 1, wherein the result report determines whether the fourth abnormality is acceptable based on predetermined criteria.

9. The abnormality inspection device according to claim 8, wherein the predetermined standard is variable.

10. The abnormality inspection apparatus according to claim 9, wherein the evaluation unit acquires data of the target image to be formed and determines the predetermined standard according to the data.

11. The abnormality inspection device according to claim 10, wherein the evaluation unit dynamically determines the predetermined standard corresponding to the position in the second read image based on the data.

12. The abnormality inspection apparatus according to claim 1, wherein in the results report, when the first abnormality and the third abnormality overlap at the location of the second abnormality, the first abnormality and the third abnormality are shown in correspondence.

13. The abnormality inspection apparatus according to claim 12, wherein the result report distinguishes between the first abnormality and the third abnormality which are associated with each other, and the second abnormality which does not overlap with the first abnormality and the first abnormality which does not overlap with the second abnormality.

14. The system includes a identifying unit that identifies the second and third abnormalities on the recording medium at locations on the recording medium that do not overlap with the first abnormality, as a fourth abnormality on the recording medium. The abnormality inspection apparatus according to claim 2, wherein the result report includes a captured image of the test image, and the test image shows a marker indicating the location where the first abnormality and the fourth abnormality were detected.

15. The system includes a identifying unit that identifies the second and third abnormalities on the recording medium at locations on the recording medium that do not overlap with the first abnormality, as a fourth abnormality on the recording medium. The specified part takes the second streak strength of the second abnormality that does not overlap with the first abnormality, and the third streak strength of the third abnormality as the fourth streak strength. The abnormality inspection apparatus according to claim 4, wherein the result report includes an indication of the first streak strength for the first abnormality and the fourth streak strength for the fourth abnormality.

16. The abnormality inspection device according to claim 8, wherein the evaluation unit has a learned model for determining the acceptable reference value for the result of the evaluation.

17. The abnormality inspection device according to claim 1, wherein the predetermined format for outputting the result report is customizable.

18. A first setting step involves detecting a streak-like first anomaly caused by the image reading unit from a first read image obtained by the image reading unit, and setting a first streak intensity for the first anomaly. A second setting step involves detecting an abnormality from the second read image of the recording medium by the image reading unit, and setting a second streak intensity for the detected streak-like second abnormality. An evaluation step to evaluate whether a third abnormality in the form of streaks has occurred at a location on the recording medium that overlaps with the first abnormality, based on the first and second streaks. Output step to output the results report of the evaluation, An abnormality detection method that includes this.

19. Computers A first setting means for detecting a streak-like first anomaly caused by the image reading unit from a first read image by the image reading unit, and for setting a first streak intensity for the first anomaly, A second setting means detects an abnormality from the second read image of the recording medium by the image reading unit and sets a second streak intensity for the detected streak-like second abnormality. An evaluation means for evaluating whether a third abnormality in the form of streaks has occurred at a location on the recording medium that overlaps with the first abnormality, based on the first and second streaks. Output means for outputting the results report of the evaluation, A program that makes it function as such.

Images