Image inspection device, image inspection method, and image inspection program

Abstract

To properly detect a specific abnormality from a read image.SOLUTION: An image inspection apparatus includes an image reading unit that reads an original image formed on a recording material on the basis of a print job and generates a read image, and a control unit that analyzes the read image and performs image inspection, and a read image is obtained from the image reading unit, an edge is detected from the read image, a region in the vicinity of the edge is excluded from the image inspection target, and a predetermined filter process is performed on the read image after the exclusion process to generate a first reference image, a difference between the read image after the exclusion process and the first reference image is extracted to generate a first comparison image, and a first comparison image is binarized by using a predetermined threshold value, and a location where a specific abnormality has occurred is detected and the detection result is output.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an image inspection device, an image inspection method, and an image inspection program, and more particularly to an image inspection device, an image inspection method, and an image inspection program for inspecting a read image. [Background technology]

[0002] Electrophotographic prints can sometimes have defects that are not present in the image being printed. For example, in electrophotography, spot-like defects caused by the so-called firefly phenomenon can occur during transfer, and these spot-like defects appear as faint undulations on halftone images. These firefly-like defects occur when developer is packed at a high density, causing the developer to compress and solidify within the container body, and these developer clumps then adhere to the paper.

[0003] There are many known methods for optically inspecting such abnormalities, and they are not limited to the field of printing. For example, Patent Document 1 listed below discloses a method for visually inspecting semiconductor wafers for defects in objects having repeating pattern areas, which includes a process for generating a reference image by excluding changes in shading due to defects from a captured image of the repeating pattern area, a process for obtaining an image in which only steep changes in shading are extracted by comparing the reference image with the captured image, and a process for binarizing the brightness value of the extracted image to determine whether or not there is a defective part. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2001-209798 Summary of the Invention [Problem to be solved by the invention]

[0005] In the method of Patent Document 1, a smoothing process is performed to remove frequency components of the defect to be detected from the captured image, thereby generating a reference image. However, in the case of printed matter, the image to be printed contains various types of content, and for example, if the image to be printed contains content with frequency components equivalent to the abnormality to be detected, detecting the defect using the method of Patent Document 1 will result in erroneous detection of content in which no abnormality has occurred.

[0006] The present invention has been made in consideration of the above-mentioned problems, and its main object is to provide an image inspection device, an image inspection method, and an image inspection program that can appropriately detect specific abnormalities from a read image. [Means for solving the problem]

[0007] According to one aspect of the present invention, in an image inspection device including an image reading unit that reads an original image formed on a recording material based on a print job and generates a read image, and a control unit that analyzes the read image and performs image inspection, the control unit includes an exclusion processing unit that acquires the read image from the image reading unit, detects edges with relatively high contrast from the read image, and excludes areas near the edges from targets for the image inspection, a filter processing unit that performs a predetermined filter process on the read image after the exclusion process to generate a first reference image, a comparison processing unit that compares the read image after the exclusion process with the first reference image to generate a first comparison image, and an abnormality location processing unit that binarizes the first comparison image using a predetermined threshold value to detect areas where a specific abnormality has occurred and outputs the detection result. A detection unit.

[0008] One aspect of the present invention is an image inspection method in an image inspection device that includes an image reading unit that reads an original image formed on a recording material based on a print job and generates a read image, and a control unit that analyzes the read image and performs image inspection, and the method performs the following steps: acquiring the read image from the image reading unit, detecting edges with relatively high contrast from the read image, and excluding areas near the edges from the image inspection; filtering that performs a predetermined filter process on the read image after the exclusion process to generate a first reference image; comparison that compares the read image after the exclusion process with the first reference image to generate a first comparison image; and abnormality detection that binarizes the first comparison image using a predetermined threshold, detects areas where a specific abnormality has occurred, and outputs the detection result.

[0009] One aspect of the present invention is an image inspection program that operates on an image inspection device that includes an image reading unit that reads an original image formed on a recording material based on a print job and generates a read image, and a control unit that analyzes the read image and performs image inspection, and causes the control unit to perform an exclusion process that acquires the read image from the image reading unit, detects edges with relatively high contrast from the read image, and excludes areas near the edges from the image inspection target, a filtering process that performs a predetermined filter process on the read image after the exclusion process to generate a first reference image, a comparison process that compares the read image after the exclusion process with the first reference image to generate a first comparison image, and an abnormality detection process that binarizes the first comparison image using a predetermined threshold, detects areas where a specific abnormality has occurred, and outputs the detection result. [Effects of the Invention]

[0010] According to the image inspection device, the image inspection method, and the image inspection program of the present invention, it is possible to appropriately detect a specific abnormality from a read image.

[0011] The reason is that in an image inspection device that includes an image reading unit that reads an original image formed on a recording material based on a print job and generates a read image, and a control unit that analyzes the read image and performs image inspection, the read image is obtained from the image reading unit, edges are detected from the read image, the area near the edge is excluded from the target of image inspection, a predetermined filter process is performed on the read image after the exclusion process to generate a first reference image, the difference between the read image after the exclusion process and the first reference image is extracted to generate a first comparison image, the first comparison image is binarized using a predetermined threshold, the location where a specific abnormality has occurred is detected, and the detection result is output. [Brief explanation of the drawings]

[0012] [Figure 1] 1A and 1B are schematic diagrams illustrating an image inspection method according to an embodiment of the present invention. [Figure 2] 1 is a schematic diagram showing an example of the configuration of an image inspection system according to a first embodiment of the present invention. [Figure 3] FIG. 3 is a schematic diagram showing another example of the configuration of the image inspection system according to the first embodiment of the present invention. [Figure 4] 1 is a block diagram showing the configuration of an image inspection device according to a first embodiment of the present invention. [Figure 5] FIG. 2 is a flowchart showing the operation of the image inspection device according to the first embodiment of the present invention. [Figure 6] FIG. 3 is a flowchart showing the operation (abnormal portion detection processing) of the image inspection device according to the first embodiment of the present invention. [Figure 7] FIG. 1 is a schematic diagram illustrating an image inspection method according to a first embodiment of the present invention. [Figure 8] FIG. 3 is a schematic diagram illustrating the effect of a threshold value in the image inspection method according to the first embodiment of the present invention. [Figure 9] FIG. 10 is a flowchart showing the operation (abnormal portion detection processing) of the image inspection device according to the second embodiment of the present invention. [Figure 10] FIG. 6 is a schematic diagram illustrating an image inspection method according to a second embodiment of the present invention. [Figure 11] FIG. 11 is a flowchart showing the operation (abnormal portion detection processing) of the image inspection device according to the third embodiment of the present invention. [Figure 12] FIG. 10 is a schematic diagram illustrating an image inspection method according to a third embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0013] As described in the background art, in electrophotography, spot-like abnormalities due to the firefly phenomenon can occur during transfer, and these spot-like abnormalities appear as faint undulations on a halftone image. As a method for optically inspecting such image abnormalities, Patent Document 1 describes a method for visually inspecting semiconductor wafers, in which a reference image is generated by smoothing a captured image. However, in the case of printed matter, the image to be printed contains various types of content, and if the image to be printed contains content with frequency components equivalent to the abnormality to be detected, the method will erroneously detect content in which no abnormality has occurred.

[0014] Therefore, in one embodiment of the present invention, in order to properly detect specific abnormalities such as those caused by the firefly phenomenon without erroneously detecting the content contained in the image to be printed (referred to as the original image), edges with relatively high contrast are detected from the read image, the area near the edge is excluded from the image inspection target, a predetermined filter process is performed on the read image after the exclusion process to generate a first reference image, the read image after the exclusion process is compared with the first reference image to generate a first comparison image, and the first comparison image is binarized using a predetermined threshold to detect the location where the specific abnormality has occurred.

[0015] Specifically, as shown in FIG. 1, when a scanned image contains both content present in the original image and an anomaly (see FIG. 1(a)), edges with relatively high contrast are detected from the scanned image using a known method (e.g., a method in which an edge is defined as a position where the absolute value of the first derivative of density change is maximum or a position where the second derivative of density change crosses zero), and the area near the edge is identified (see FIG. 1(b)). Next, an image is generated in which the identified edge area is excluded from the detection target (see FIG. 1(c)). A predetermined filter process (e.g., blurring) is performed on the image to generate a reference image (see FIG. 1(d)). Next, the image in which the edge area is excluded is compared with the reference image (i.e., the difference is extracted) to generate a comparison image (see FIG. 1(e)). The comparison image is then binarized using a predetermined threshold to detect the anomaly (see FIG. 1(f)). For example, abnormal areas are detected by binarizing the image using a threshold value obtained from a comparison image generated by performing the same processing on the original image, or abnormal areas detected by binarizing the image using a threshold value are narrowed down using an area threshold value or a circularity threshold value determined according to the specific abnormality.

[0016] In this way, by excluding areas near edges with relatively high contrast from the detection target for anomalies, it is possible to prevent erroneous detection of content such as text and graphics contained in the original image. Furthermore, by performing a blurring process as a filter process, it is possible to include anomalies caused by the firefly phenomenon in the comparison image, thereby enabling reliable detection of specific anomalies. Furthermore, by using a threshold value determined from a comparison image generated from the original image as the threshold value, it is possible to prevent erroneous detection of content contained in the original image, and by using an area threshold value or a circularity threshold value as the threshold value, it is possible to reliably detect specific anomalies. [Example]

[0017] To explain the above-described embodiment of the present invention in more detail, the first embodiment of the present invention will be described. An image inspection device, an image inspection method, and an image inspection program according to an embodiment will be described with reference to Figs. 2 to 8. Figs. 2 and 3 are schematic diagrams showing an example of the configuration of an image inspection system according to this embodiment, and Fig. 4 is a block diagram showing the configuration of the image inspection device according to this embodiment. Figs. 5 and 6 are flow charts showing the operation of the image inspection device according to this embodiment, Fig. 7 is a schematic diagram explaining the image inspection method according to this embodiment, and Fig. 8 is a schematic diagram explaining the effect of a threshold value in the image inspection method according to this embodiment.

[0018] As shown in FIG. 2, the image inspection system 10 of this embodiment performs RIP (Raster Image Processing) on a print job input from a computer device (not shown). The image forming device 12 includes a controller 11 having a RIP unit that outputs image data after RIP processing (called original image data), an image forming device 12 having a printing unit that forms an image (original image) on a recording material (paper) based on the original image data, and an image inspection device 20 that reads the original image formed on the recording material and inspects the finish of the print. These are connected to a communication network such as a LAN (Local Area Network) or WAN (Wide Area Network) defined by standards such as Ethernet (registered trademark), token ring, and FDDI (Fiber-Distributed Data Interface). The devices are connected via a network 13 so as to be able to communicate data.

[0019] The RIP unit of the controller 11 translates a print job written in a PDL (Page Description Language) such as PCL (Printer Control Language) or PS (Post Script) to generate intermediate data, performs color conversion on the intermediate data using a color conversion table, and performs rendering to generate image data for each page.

[0020] The printing unit of the image forming device 12 is composed of components necessary for image formation using an imaging process such as electrophotography, and prints an image based on image data onto designated paper. Specifically, an electrostatic latent image is formed by irradiating a photosensitive drum charged by a charging device with light corresponding to the image from an exposure device, and then a developing device develops the image by attaching charged toner to the photosensitive drum. The toner image is then primarily transferred to a transfer belt, and secondarily transferred from the transfer belt to paper, and the toner image on the paper is then fixed by a fixing device.

[0021] 2, the image inspection system 10 is configured with the controller 11, the image forming device 12, and the image inspection device 20, but if the image forming device 12 is given the function of the image inspection device 20, the image inspection system 10 can also be configured with the controller 11 and the image forming device 12, as shown in FIG. 3. Also, if the image forming device 12 is given the function of the controller 11 (if a RIP unit is provided in the image forming device 12), the image inspection system 10 can also be configured with the image forming device 12 and the image inspection device 20 (or only the image inspection device 20). Below, the image inspection device 20 will be described based on the configuration in FIG. 2.

[0022] The image inspection device 20 in the configuration of Figure 2 is composed of a control unit 21, a memory unit 22, a network I / F unit 23, a display operation unit 24, a paper feed unit 25, an image reading unit 26, a paper discharge unit 27, etc., as shown in Figure 4(a).

[0023] The control unit 21 is composed of a CPU (Central Processing Unit) 21a and memories such as a ROM (Read Only Memory) 21b and a RAM (Random Access Memory) 21c. The CPU 21a reads out a program from the ROM 21b or the storage unit 22, loads it into the RAM 21c, and executes it, thereby controlling the entire image inspection device 20.

[0024] The storage unit 22 is composed of a hard disk drive (HDD) or a solid state drive (SSD), and stores programs for the CPU 21a to control each unit, original image data, scanned image data obtained by scanning an original image formed on paper, threshold values calculated based on the original image, etc. do.

[0025] The network I / F unit 23 is composed of a NIC (Network Interface Card), a modem, etc., and connects the image inspection device 20 to the communication network 13 so that the original image data can be received from the controller 11 .

[0026] The display operation unit 24 is configured with a touch panel or the like on which an operation unit such as a touch sensor with electrodes arranged in a grid pattern is formed on a display unit such as an LCD (Liquid Crystal Display) or an organic EL (Electro Luminescence) display, and displays various screens related to the operation of the image inspection device 20 and accepts various operations related to the operation of the image inspection device 20. Note that the operation unit may be equipped with hard keys or the display unit and the operation unit may be separate devices.

[0027] The paper feed unit 25 is configured with one or more paper feed trays, and conveys paper on which an image (original image) has been formed by the image forming device 12 to the image reading unit .

[0028] The image reading unit 26 is a part that reads an image (original image) formed on paper, and for example, optically scans the paper on which the image is formed, and reads the image by focusing the reflected light from the paper on the light receiving surface of a sensor such as a CCD (Charge Coupled Device).

[0029] The paper discharge unit 27 is composed of one or more paper discharge trays and discharges the paper after the image has been read. In this embodiment, it is preferable to provide multiple paper discharge trays so that paper on which a specific abnormality (in this embodiment, an abnormality caused by the firefly phenomenon) has been detected by the abnormality portion detection unit 31 (described later) can be distinguished from normal paper and discharged.

[0030] As shown in FIG. 4(b), the control unit 21 also functions as an exclusion processing unit 28, a filter processing unit 29, a comparison processing unit 30, an abnormality portion detection unit 31, and the like.

[0031] The exclusion processing unit 28 acquires a read image from the image reading unit 26, detects edges with relatively high contrast from the read image, and excludes areas near the edges (areas surrounding a series of edges) from the target of image inspection (detection of abnormalities), so that content such as text or graphics included in the original image is not erroneously detected as an abnormality. The exclusion processing unit 28 also acquires an original image from the controller 11, the image forming device 12, or the like, detects edges with relatively high contrast from the original image, and excludes areas near the edges from the target of detection of abnormalities. If the original image contains content having frequency components equivalent to a specific abnormality (for example, an abnormality caused by the firefly phenomenon), the exclusion processing unit 28 excludes the content from the target of abnormality detection, as necessary. Note that exclusion includes both cases where the area near the edge or content is deleted from the scanned image or original image (if the image has a uniform background, the area near the edge or content is overwritten with the background image), and cases where the area near the edge or content is not deleted (or overwritten with the background image) from the scanned image or original image, but rather set so that abnormalities are not detected in the area near the edge or content.

[0032] The filter processing unit 29 performs a filter process (blurring process, for example, a process of averaging the pixel value of a pixel of interest using the pixel values of its surrounding pixels) on the read image after the exclusion process by the exclusion processing unit 28 to filter out specific abnormalities (for example, abnormalities caused by the firefly phenomenon), thereby generating a reference image (first reference image). Furthermore, the filter processing unit 29 performs a filter process similar to that on the read image on the original image after the exclusion process by the exclusion processing unit 28, thereby generating a reference image (second reference image).

[0033] The comparison processing unit 30 compares the read image after the exclusion processing by the exclusion processing unit 28 with the reference image. The comparison processing unit 30 compares the original image after the exclusion processing by the exclusion processing unit 28 with the reference image (second reference image) to extract the difference, and generates a comparison image (second comparison image) consisting of the difference.

[0034] The abnormal portion detection unit 31 binarizes the first comparison image using a predetermined threshold value, detects a portion where a specific abnormality (for example, an abnormality caused by the firefly phenomenon) has occurred, and outputs the detection result. In this embodiment, in particular, a threshold value is calculated based on the second comparison image (a comparison image generated from the original image), and the first comparison image is binarized using the threshold value to detect a portion where a specific abnormality has occurred, and outputs the detection result. For example, the detection result may be displayed on the display operation unit 24, or a sheet of paper where a specific abnormality has been detected may be output to a paper output tray different from the paper output tray for normal sheets.

[0035] The above-mentioned exclusion processing unit 28, filter processing unit 29, comparison processing unit 30, and abnormal part detection unit 31 may be configured as hardware, or the control unit 21 may be configured as an image inspection program that causes the control unit 21 to function as the exclusion processing unit 28, filter processing unit 29, comparison processing unit 30, and abnormal part detection unit 31, and the image inspection program may be executed by the CPU 21a.

[0036] 4 shows an example of the image inspection device 20 of this embodiment, and its configuration can be changed as appropriate. For example, in the case of the image inspection system 10 configured as shown in FIG. 3, an image formed on paper can be read using an inline sensor of the image forming device 12, and in that case, the paper feed unit 25, the image reading unit 26, and the paper discharge unit 27 can be omitted.

[0037] Specific operations of the image inspection device 20 of this embodiment will be described below with reference to Figures 5 to 8. The CPU 21a executes the processing of each step shown in the flowcharts of Figures 5 and 6 by loading an image inspection program stored in the ROM 21b or the storage unit 22 into the RAM 21c and executing it. In the following description, it is assumed that the original image is a uniform halftone image (represented by dot hatching in the figure) as shown in Figure 7, which includes one spot-like content and one text content (see Figure 7(f)), and that a spot-like abnormality has occurred in one location in the read image obtained by reading the original image (see Figure 7(a)).

[0038] The control unit 21 (exclusion processing unit 28) acquires a scanned image from the image reader 26, detects edges with relatively high contrast from the scanned image, identifies areas near the edges, and excludes the areas near the identified edges from the detection target for abnormalities (S101). For example, as shown in FIG. 7(b), text areas are identified by detecting characters printed in a color (e.g., black) that contrasts highly with the background color, and the identified areas are removed from the scanned image as shown in FIG. 7(c). Because gradation changes sharply near edges, there is a risk of false detection when detecting abnormalities. Therefore, false detection can be prevented by detecting edge portions and excluding their vicinity from the detection target for abnormalities. A commonly known method for detecting edges is to use a differential filter, such as a Sobel filter. The position where the absolute value of the first derivative of the density change is maximum or the position where the second derivative of the density change crosses zero can be detected as an edge.

[0039] Next, the control unit 21 (filter processing unit 29) performs a predetermined filter process on the read image after the exclusion process to generate a reference image (S102, see FIG. 7(d)). This filter is intended to remove frequencies that belong to the abnormality to be detected, and the filter strength and size depend on the abnormality to be detected. For example, since the firefly phenomenon often occurs as a phenomenon with a diameter of about 1 to 3 mm, a reference image in which the influence of the firefly phenomenon has been removed can be generated by setting the filter size to, for example, about 5 mm.

[0040] Next, the control unit 21 (comparison processing unit 30) compares the read image with the reference image to generate a comparison image (S103, see FIG. 7(e)). For example, the comparison image is generated by calculating the difference in pixel values at corresponding positions in the two images. When this difference is calculated, a difference value is generated only at the location where there is a difference between the two images, that is, only at the location where a spot-like abnormality originally occurred.

[0041] Next, the control unit 21 (abnormal part detection unit 31) processes (binarizes) the comparison image using a predetermined threshold value, detects abnormal parts, and outputs the detection results (S104, see FIG. 7(k)). Since difference values occur in parts where spot-like abnormalities have occurred in the comparison image, this processing makes it possible to detect the parts where abnormalities have occurred.

[0042] In many cases, the original image contains not only halftones but also some other content. For example, if the original image contains content with the same frequency components as the anomaly you want to detect, detecting the anomaly using only the scanned image will result in false positives in areas where the content does not have an anomaly.

[0043] Therefore, in this embodiment, in areas where content exists in the original image, the threshold for detecting abnormal areas is increased and the detection sensitivity is reduced to prevent erroneous detection of content areas. Specifically, the original image is processed in the same way as the scanned image, a comparison image is generated from the original image, and the threshold is set using the comparison image. Figure 6 shows a specific example of the abnormal area detection process of S104 in Figure 5.

[0044] First, the control unit 21 (exclusion processing unit 28) acquires an original image from the controller 11, the image forming device 12, or the like, detects high-contrast edges from the original image, identifies areas near the edges, and excludes the areas near the identified edges from detection targets for abnormal portions (S201, see FIGS. 7(f) to (h)). Next, the control unit 21 (filter processing unit 29) performs the above-mentioned predetermined filter processing on the original image after the exclusion processing to generate a reference image (S202, see FIG. 7(i)). Next, the control unit 21 (comparison processing unit 30) compares the original image with the reference image to generate a comparison image (S203, see FIG. 7(j)). In this image, difference values occur only in areas of content that have frequency components equivalent to spot-like abnormalities present in the original image.

[0045] Next, the control unit 21 (abnormal part detection unit 31) sets a threshold value for each position using the comparison image obtained from the original image (S204), processes (binarizes) the comparison image generated from the read image using the set threshold value to detect abnormal parts, and outputs the detection result (S205, see Figure 7(k)).

[0046] Figure 8 shows the effect of a threshold set based on a comparison image obtained from an original image. In Figure 8, points A and B are areas where content exists in the original image, and point C is an area where an abnormality has occurred. In this case, if the entire image is binarized using a fixed threshold, point B will be detected in addition to point C because the pixel values of points B and C exceed the threshold. Therefore, the threshold is changed for each position depending on the comparison image of the original image. In other words, because points A and B have difference values even in the comparison image obtained from the original image, the threshold can be raised to reduce detection sensitivity. On the other hand, point C does not have a difference value in the comparison image obtained from the original image, so the threshold does not need to be raised and the abnormality can be detected.

[0047] As explained above, by excluding areas near edges with relatively high contrast from the detection target for abnormalities, it is possible to prevent erroneous detection of content such as text and graphics contained in the original image. This allows the comparison image to include anomalies caused by the firefly phenomenon, making it possible to reliably detect specific anomalies. Furthermore, by using a threshold value determined from a comparison image generated from an original image, it is possible to prevent erroneous detection of content contained in the original image. [Example]

[0048] Next, an image inspection device, an image inspection method, and an image inspection program according to a second embodiment of the present invention will be described with reference to Fig. 9 and Fig. 10. Fig. 9 is a flowchart showing the operation (abnormal portion detection processing) of the image inspection device of this embodiment, and Fig. 10 is a schematic diagram explaining the image inspection method of this embodiment.

[0049] In the first embodiment described above, the abnormality detection process involves setting a threshold value using a comparison image generated from the original image, and processing the comparison image generated from the scanned image with that threshold value to detect abnormalities, but even when detecting abnormalities using this threshold value, false positives may occur. This is due to noise with minute peaks that are not actually visible, or large irregularities that are not the target of detection.

[0050] To avoid such erroneous detection, in this embodiment, the detected locations are narrowed down based on their area. In this case, the configuration of the image inspection device 20 is the same as that of the first embodiment described above, but the control unit 21 (abnormal location detection unit 31) separates an area for each detected abnormal location, calculates the area of each area, and excludes from the detection results any abnormal location whose calculated area deviates from a threshold set according to a specific abnormality (for example, an abnormality based on the firefly phenomenon), thereby narrowing down the abnormal locations.

[0051] Specific operations of the image inspection device 20 of this embodiment will be described below with reference to Fig. 9 and Fig. 10. The CPU 21a loads an image inspection program stored in the ROM 21b or the storage unit 22 into the RAM 21c and executes it, thereby executing the processing of each step shown in the flowchart of Fig. 9. Note that Fig. 9 shows a specific example of the abnormal portion detection processing of S104 in Fig. 5 of the first embodiment.

[0052] First, as in the first embodiment, the control unit 21 (exclusion processing unit 28) detects edges with relatively high contrast from the original image, identifies areas near the edges, and excludes the areas near the identified edges from detection targets for abnormal regions (S301). Next, the control unit 21 (filter processing unit 29) performs a predetermined filter process on the original image after the exclusion process to generate a reference image (S302). Next, the control unit 21 (comparison processing unit 30) compares the original image with the reference image to generate a comparison image (S303). Next, the control unit 21 (abnormal region detection unit 31) sets a threshold value for each position using the comparison image obtained from the original image (S304), and processes (binarizes) the comparison image generated from the scanned image using the set threshold value to detect abnormal regions (S305, see FIG. 10(k)).

[0053] Next, the control unit 21 (abnormal part detection unit 31) separates areas for each detected part (S306, see FIG. 10(l)), calculates the area of each area (S307, see FIG. 10(m)), and excludes from the detection results any abnormal part whose calculated area does not fall within a threshold set according to a specific abnormality (for example, an abnormality caused by the firefly phenomenon), thereby narrowing down the abnormal parts (S308, see FIG. 10(n)).

[0054] For example, if the anomaly to be detected is a spot-like anomaly caused by the firefly phenomenon, it is likely to occur as an anomaly with a diameter of about 1 to 3 mm because the cause is a core foreign object. Therefore, by setting an area corresponding to a diameter of 1 to 3 mm as a threshold and excluding areas whose area falls outside the threshold, it is possible to prevent erroneous detection of anomalies other than those caused by the firefly phenomenon.

[0055] In this case, the abnormality is detected using a threshold value set based on the comparison image obtained from the original image. After detecting the abnormality, the abnormality is narrowed down using an area threshold. However, if the original image does not contain content having frequency components equivalent to the abnormality to be detected, the abnormality may be detected using a predetermined threshold (for example, a fixed threshold), and then the abnormality may be narrowed down using an area threshold.

[0056] As described above, by narrowing down the abnormal locations using a threshold value (area threshold value) set according to the abnormality to be detected, it is possible to reliably detect a specific abnormality. [Example]

[0057] Next, an image inspection device, an image inspection method, and an image inspection program according to a third embodiment of the present invention will be described with reference to Fig. 11 and Fig. 12. Fig. 11 is a flowchart showing the operation of the image inspection device of this embodiment, and Fig. 12 is a schematic diagram explaining the image inspection method of this embodiment.

[0058] In the second embodiment described above, an example was shown in which a detected location was excluded based on its area, but false detections can occur when an abnormality with a different shape than the abnormality desired to be detected is detected. For example, if the abnormality desired to be detected is a spot-like abnormality caused by the firefly phenomenon, an optically-induced streak or the like may be falsely detected.

[0059] To avoid such erroneous detection, in this embodiment, the detected locations are narrowed down based on the circularity. In this case, the configuration of the image inspection device 20 is the same as that of the first embodiment described above, but the control unit 21 (abnormal location detection unit 31) separates an area for each detected abnormal location, calculates the circularity of each area, and excludes from the detection results any abnormal location whose calculated circularity deviates from a threshold set according to a specific abnormality (for example, an abnormality caused by the firefly phenomenon), thereby narrowing down the abnormal locations.

[0060] Specific operations of the image inspection device 20 of this embodiment will be described below with reference to Fig. 11 and Fig. 12. The CPU 21a loads an image inspection program stored in the ROM 21b or the storage unit 22 into the RAM 21c and executes it, thereby executing the processing of each step shown in the flowchart of Fig. 11. Note that Fig. 11 shows a specific example of the abnormal portion detection processing of S104 in Fig. 5 of the first embodiment.

[0061] First, as in the first embodiment, the control unit 21 (exclusion processing unit 28) detects edges with relatively high contrast from the original image, identifies areas near the edges, and excludes the areas near the identified edges from detection targets for abnormal areas (S401). Next, the control unit 21 (filter processing unit 29) performs a predetermined filter process on the original image after the exclusion process to generate a reference image (S402). Next, the control unit 21 (comparison processing unit 30) compares the original image with the reference image to generate a comparison image (S403). Next, the control unit 21 (abnormal area detection unit 31) sets a threshold value for each position using the comparison image obtained from the original image (S404), and processes (binarizes) the comparison image generated from the scanned image using the set threshold value to detect abnormal areas (S405, see FIG. 12(k)).

[0062] Next, the control unit 21 (abnormal portion detection unit 31) separates the regions for each detected portion (S406, see FIG. 12(l)), calculates the circularity of each region (S407, see FIG. 12(m)), and excludes from the detection results any abnormal portion whose calculated circularity does not fall within a threshold set according to a specific abnormality (for example, an abnormality caused by the firefly phenomenon), thereby narrowing down the abnormal portions (S408, see FIG. 10(n)). The circularity is the ratio of the area of the actual region to the area of a circle calculated at the maximum distance from the center pixel of the region (the ratio of the area of the region to the area of the circle, a value between 0 and 1), and is 1 if the region is a perfect circle.

[0063] For example, if the abnormality you want to detect is a spot-like abnormality caused by the firefly phenomenon, Considering this, the range of influence of the firefly phenomenon is close to circular. Therefore, by taking this characteristic into consideration, setting the circularity as a threshold and excluding areas where the circularity is outside the threshold (for example, areas with a circularity of 0.5 or less), it is possible to prevent erroneous detection of abnormalities other than those caused by the firefly phenomenon.

[0064] Here, abnormal areas are detected using a threshold value set based on a comparison image obtained from the original image, and then the abnormal areas are narrowed down using a circularity threshold value. However, if the original image does not contain content with frequency components equivalent to the abnormality to be detected, abnormal areas may be detected using a predetermined threshold value (for example, a fixed threshold value), and then the abnormal areas may be narrowed down using a circularity threshold value.

[0065] As described above, by narrowing down the abnormal locations using a threshold value (threshold value of circularity) set according to the abnormality to be detected, it is possible to reliably detect a specific abnormality.

[0066] The present invention is not limited to the above-described embodiments, and the configuration and control can be modified as appropriate without departing from the spirit of the present invention.

[0067] For example, the first embodiment describes a case where a threshold value set based on a comparison image obtained from an original image is used, the second embodiment describes a case where a threshold value for area set according to the abnormality to be detected is used, and the third embodiment describes a case where a threshold value for circularity set according to the abnormality to be detected is used, but these can be combined in any way. [Industrial Applicability]

[0068] The present invention can be used in an image inspection device that inspects a read image, an image inspection method in the image inspection device, an image inspection program that runs in the image inspection device, and a recording medium on which the image inspection program is recorded. [Explanation of symbols]

[0069] 10. Image inspection system 11 Controller 12 Image forming device 13. Communication Networks 20 Image inspection equipment 21 Control Unit 21a CPU 21b ROM 21c RAM 22 Memory section 23 Network I / F section 24 Display operation section 25 Paper feed section 26 Image reading unit 27 Paper output section 28 Exclusion Processing Unit 29 Filter processing section 30 Comparison processing section 31 Abnormality detection unit

Claims

1. An image inspection device comprising a control unit that acquires a read image obtained by reading an image formed on a recording material, compares the read image with a reference image, and, for areas from which differences are extracted by the comparison, detects areas in the read image where the circularity, which indicates closeness to a perfect circle, exceeds a predetermined value as an abnormality.

2. 2. The image inspection device according to claim 1, wherein the reference image is an image generated by performing a predetermined filter process on the scanned image.

3. 3. The image inspection device according to claim 1, wherein the reference image is an image generated by performing a blurring process on the scanned image.

4. The image inspection device according to any one of claims 1 to 3, wherein the control unit detects the abnormality by binarizing difference information extracted based on a comparison between the read image and the reference image using a predetermined threshold value.

5. The image inspection device according to claim 4 , wherein the control unit generates a comparison image based on the difference information and detects the abnormality by binarizing the comparison image.

6. the control unit detects edges from the read image and excludes areas near the edges from the detection target for the abnormality; 6. An image inspection device according to claim 1, wherein the scanned image excluding the area near the edge is compared with the reference image generated based on the scanned image excluding the area near the edge.

7. The image inspection device according to claim 4 , wherein the control unit sets the threshold value using a second reference image generated based on the original image formed on the recording material.

8. 8. The image inspection device according to claim 7, wherein the second reference image is an image generated by performing a predetermined filtering process on the original image.

9. 9. The image inspection device according to claim 7, wherein the second reference image is an image generated by performing a blurring process on the original image.

10. 10. An image inspection device according to claim 7, wherein the control unit sets the threshold value based on difference information extracted based on a comparison between the original image and the second reference image.

11. 11. The image inspection device according to claim 1, wherein the abnormality is an abnormality based on the firefly phenomenon.

12. 12. The image inspection device according to claim 1, wherein the circularity is set based on an original image formed on the recording material.

13. An image inspection program executed by a computer of an image inspection device, A step of acquiring a read image by reading an image formed on a recording material; comparing the read image with a reference image, and detecting, among areas where a difference is detected by the comparison, areas where a circularity indicating closeness to a perfect circle exceeds a predetermined value as an abnormality in the read image; An image inspection program for causing the computer to execute the above.

14. 14. The image inspection program according to claim 13, wherein the reference image is an image generated by performing a predetermined filter process on the scanned image.

15. 15. The image inspection program according to claim 13, wherein the reference image is an image generated by performing a blurring process on the scanned image.

16. The image inspection program according to any one of claims 13 to 15, wherein the step of detecting the abnormality detects the abnormality by binarizing difference information extracted based on a comparison between the read image and the reference image using a predetermined threshold value.

17. 17. The image inspection program according to claim 16, wherein the step of detecting an abnormality comprises generating a comparison image based on the difference information and binarizing the comparison image to detect the abnormality.

18. The computer, further executing a step of detecting an edge from the read image and excluding an area near the edge from the detection target for the abnormality; The image inspection program according to any one of claims 13 to 17, wherein the comparing step compares the read image excluding the area near the edge with the reference image generated based on the read image excluding the area near the edge.

19. The image inspection program according to claim 16, wherein the threshold value is set using a second reference image generated based on the original image formed on the recording material.

20. 20. The image inspection program according to claim 19, wherein the second reference image is an image generated by performing a predetermined filter process on the original image.

21. 21. The image inspection program according to claim 19, wherein the second reference image is an image generated by performing a blurring process on the original image.

22. 22. The image inspection program according to claim 19, wherein the threshold is set based on difference information extracted based on a comparison between the original image and the second reference image.

23. The image inspection program according to any one of claims 13 to 22, wherein the abnormality is an abnormality based on the firefly phenomenon.

24. 24. The image inspection program according to claim 13, wherein the circularity is set based on an original image formed on the recording material.

25. An image inspection method performed by an image inspection device, comprising: A step of acquiring a read image by reading an image formed on a recording material; comparing the read image with a reference image, and detecting, among areas where a difference is detected by the comparison, areas where a circularity indicating closeness to a perfect circle exceeds a predetermined value as an abnormality in the read image; An image inspection method comprising:

26. 26. The image inspection method according to claim 25, wherein the circularity is set based on an original image formed on the recording material.

Images