Image processing method and image processing device

Abstract

The present invention provides an image processing method and an image processing device, the purpose of which is to extract image defects by type from a test image through simple processing. A processor (80) performs a main filtering process using the horizontal and vertical directions of the test image as processing directions, thereby generating a first pre-processed image and a second pre-processed image. The main filtering process is a process of transforming the pixel values of the focus pixels selected sequentially from the test image into a transformation value obtained by emphasizing the difference between the pixel value of the focus area and the pixel values of two adjacent areas. The processor (80) extracts a first unique portion that exists in the first pre-processed image and is not common to the two images, a second unique portion that exists in the second pre-processed image and is not common to the two images, and a third unique portion that is common to the two images as the image defects.

Description

Technical Field

[0001] The present invention relates to an image processing method and an image processing device for determining the cause of image failure based on a test image. Background Art

[0002] Image forming apparatuses such as printers and multifunction peripherals execute a print process to form an image on a sheet. During the print process, the image formed on the output sheet may have image defects such as vertical lines, horizontal lines, noise, or uneven density.

[0003] For example, if the image forming apparatus is an apparatus that performs the printing process using an electrophotographic method, the cause of the image failure may be various parts such as the photoreceptor, the charging unit, the developing unit, and the transfer unit. Furthermore, determining the cause of the image failure requires expertise.

[0004] It is also known that in an image processing device, the phenomenon of vertical lines as an example of image defects is associated with characteristic information such as the color, concentration or screen count of the vertical lines in advance as table data, and the phenomenon causing the vertical lines is determined based on the information on the color, concentration or screen count of the image of the vertical lines in the test image and the table data (for example, refer to patent document 1).

[0005] The table data is data obtained by setting ranges of parameters such as color, density, or screen ruling of an image using a threshold value for each type of phenomenon causing the vertical streaks.

[0006] Prior art literature

[0007] Patent Literature

[0008] Patent Document 1: Japanese Patent Application Laid-Open No. 2017-83544 Summary of the Invention

[0009] Technical Problems to be Solved by the Invention

[0010] However, the test image may contain multiple types of image defects. In this case, in order to simplify the determination process and improve the determination accuracy, it is preferable to determine the cause of the image defect in the test image for each type of image defect.

[0011] Furthermore, in the image processing device, it is preferable that the image defective portion can be extracted by type from the test image through simple processing.

[0012] An object of the present invention is to provide an image processing method and an image processing apparatus capable of extracting image defects occurring in an image forming apparatus by type from a test image through simple processing.

[0013] Technical solutions to technical problems

[0014] An image processing method according to one aspect of the present invention is a method in which a processor determines image defects in a test image obtained by image reading processing of a test image obtained from an output sheet of an image forming device. The image processing method includes the processor performing a first pre-processing including a main filtering process with the horizontal direction of the test image as the processing direction, thereby generating a first pre-processed image. The main filtering process converts the pixel values of pixels of interest selected sequentially from the test image into conversion values. The conversion values are obtained by emphasizing the difference between the pixel values of a region of interest containing the pixel of interest and the pixel values of two adjacent regions on either side of the region of interest along a predetermined processing direction. Furthermore, the image processing method includes the processor performing a second pre-processing including the main filtering process with the vertical direction of the test image as the processing direction, thereby generating a second pre-processed image. Furthermore, the image processing method includes the processor performing a unique portion extraction process, wherein the unique portion extraction process extracts, as the image defects, a first unique portion, a second unique portion, and a third unique portion from among the unique portions consisting of one or more significant pixels in the first and second pre-processed images. The first unique portion is present in the first pre-processed image and is unique to the first and second pre-processed images. The second unique portion is present in the second pre-processed image and is unique to the first and second pre-processed images. The third unique portion is common to the first and second pre-processed images.

[0015] An image processing device according to another aspect of the present invention includes a processor that executes the processing of the image processing method.

[0016] Effects of the Invention

[0017] According to the present invention, it is possible to provide an image processing method and an image processing apparatus capable of extracting image defects occurring in an image forming apparatus by type from a test image through simple processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a block diagram of an image processing device according to an embodiment.

[0019] Figure 2 This is a block diagram showing the configuration of a data processing unit of the image processing apparatus according to the embodiment.

[0020] Figure 3 This is a flowchart showing an example of the procedure of image failure determination processing by the image processing apparatus according to the embodiment.

[0021] Figure 4 This is a flowchart showing an example of a procedure of a specific failure determination process performed by the image processing apparatus according to the embodiment.

[0022] Figure 5 This is a flowchart showing an example of a procedure for density unevenness determination processing by the image processing apparatus according to the embodiment.

[0023] Figure 6 A diagram showing an example of a test image including a unique portion, and an example of a pre-processed image and a feature image generated based on the test image.

[0024] Figure 7 This is a diagram showing an example of a target region and adjacent regions sequentially selected from a test image in a main filter process of the image processing device according to the embodiment.

[0025] Figure 8 This is a diagram showing an example of a test image including periodic density unevenness and longitudinal waveform data derived from the test image.

[0026] Figure 9 This is a flowchart showing an example of the procedure of a feature image generation process according to the first application example of the image processing device of the embodiment.

[0027] Figure 10 This is a flowchart showing an example of the procedure of a feature image generation process according to the second application example of the image processing device of the embodiment.

[0028] Figure 11 This is a flowchart showing an example of the procedure of a feature image generation process according to a third application example of the image processing device according to the embodiment.

[0029] Figure 12 This is a flowchart showing an example of the procedure of a feature image generation process according to a fourth application example of the image processing device according to the embodiment. DETAILED DESCRIPTION

[0030] Hereinafter, embodiments of the present invention will be described with reference to the drawings. Note that the following embodiments are examples of specific implementations of the present invention and do not limit the technical scope of the present invention.

[0031] [Configuration of Image Processing Device 10]

[0032] The image processing apparatus 10 of the embodiment includes an image forming apparatus 2 that executes a print process. The print process is a process of forming an image on a sheet. The sheet is an image forming medium such as paper or a thin sheet-like resin member.

[0033] The image processing apparatus 10 also includes the image reading device 1 that executes a reading process for reading an image from a document. For example, the image processing apparatus 10 is a copy machine, a facsimile machine, or a multifunction peripheral.

[0034] The image to be printed is an image read from the document by the image reading device 1 or an image represented by print data received from a host device (not shown). The host device is an information processing device such as a personal computer or a portable information terminal.

[0035] In addition, the image forming apparatus 2 may also form a predetermined original test image g01 on the sheet by the printing process (see Figure 6 The original test image g01 is an image that serves as a basis for the test image g1 used to determine the presence or absence of image defects and the cause of the defects in the image forming apparatus 2 (see Figure 6 ). The test image g1 will be described later.

[0036] The process including the reading process by the image reading device 1 and the printing process by the image forming device 2 based on the image obtained by the reading process is a copy process.

[0037] like Figure 1 As shown, the image forming apparatus 2 includes a sheet conveying mechanism 3 and a printing unit 4 . The sheet conveying mechanism 3 includes a sheet delivery mechanism 31 and a plurality of sheet conveying roller pairs 32 .

[0038] The sheet delivery mechanism 31 delivers the sheet from the sheet storage portion 21 to the sheet conveyance path 30 . The plurality of sheet conveyance roller pairs 32 convey the sheet along the sheet conveyance path 30 and discharge the image-formed sheet to the discharge tray 22 .

[0039] The printing unit 4 executes the printing process on the sheet conveyed by the sheet conveying mechanism 3. In this embodiment, the printing unit 4 executes the printing process using an electrophotographic method.

[0040] The printing unit 4 includes an image forming unit 4x, a laser scanning unit 4y, a transfer device 44, and a fixing device 46. The image forming unit 4x includes a drum-shaped photoreceptor 41, a charging device 42, a developing device 43, and a drum cleaning device 45.

[0041] In each image forming unit 4x, a photoreceptor 41 rotates, and a charging device 42 uniformly charges the surface of the photoreceptor 41. The charging device 42 includes a charging roller 42a that rotates in contact with the surface of the photoreceptor 41. The laser scanning unit 4y writes an electrostatic latent image on the charged surface of the photoreceptor 41 by scanning with laser light.

[0042] The developing device 43 develops the electrostatic latent image into a toner image. The developing device 43 includes a developing roller 43a that supplies toner to the photoreceptor 41. The transfer device 44 transfers the toner image on the surface of the photoreceptor 41 to the sheet. The toner is an example of a granular developer.

[0043] The fixing device 46 heats the toner image on the sheet to fix the toner image on the sheet. The fixing device 46 includes a fixing rotator 46 a that rotates in contact with the sheet and a fixing heater 46 b that heats the fixing rotator 46 a.

[0044] Figure 1 The image forming apparatus 2 shown is a tandem color printer capable of executing the aforementioned printing process for a color image. Therefore, the printing unit 4 includes four image forming units 4x corresponding to toners of different colors.

[0045] In the tandem image forming apparatus 2 , the transfer device 44 includes four primary transfer rollers 441 corresponding to the four photoreceptors 41 , an intermediate transfer belt 440 , a secondary transfer roller 442 , and a belt cleaning device 443 .

[0046] The four image forming sections 4x form the aforementioned toner images of cyan, magenta, yellow, and black, respectively, on the surface of the photoreceptor 41. Each primary transfer roller 441 is also a part of each image forming section 4x.

[0047] In each image forming unit 4x, a primary transfer roller 441 rotates while applying force to the surface of the intermediate transfer belt 440 toward the photoreceptor 41. The primary transfer roller 441 transfers the toner image from the photoreceptor 41 to the intermediate transfer belt 440. This forms a color image composed of the four-color toner images on the intermediate transfer belt 440.

[0048] In each image forming unit 4 x , the drum cleaning device 45 removes and recovers the toner remaining on the photoreceptor 41 without being transferred to the intermediate transfer belt 440 .

[0049] Secondary transfer roller 442 transfers the four-color toner image on intermediate transfer belt 440 to the sheet. In image processing apparatus 10, photoreceptor 41 and intermediate transfer belt 440 of transfer device 44 are examples of image carriers that rotate while carrying the toner image.

[0050] The belt cleaning device 443 removes and recovers the toner remaining on the intermediate transfer belt 440 without being transferred to the sheet.

[0051] like Figure 1As shown, the image processing apparatus 10 includes, in addition to the image forming apparatus 2 and the image reading apparatus 1 , a data processing unit 8 and a human-machine interface device 800 . The human-machine interface device 800 includes an operation unit 801 and a display unit 802 .

[0052] The data processing unit 8 performs various data processing related to the printing process or the reading process, and controls various electrical devices.

[0053] The operation unit 801 is a device that receives user operations. For example, the operation unit 801 includes one or both of buttons and a touch panel. The display unit 802 includes a display panel that displays information provided to the user.

[0054] like Figure 2 As shown, the data processing unit 8 includes a CPU (Central Processing Unit) 80 , a RAM (Random Access Memory) 81 , an auxiliary storage device 82 , and a communication device 83 .

[0055] The CPU 80 can process data received by the communication device 83, perform various image processing operations, and control the image forming device 2. The received data may include the print data. The CPU 80 is an example of a processor that performs data processing including the image processing. Alternatively, the CPU 80 may be implemented as another type of processor, such as a DSP (Digital Signal Processor).

[0056] The communication device 83 is a communication interface device for communicating with other devices such as the host device via a network such as a LAN (Local Area Network). The CPU 80 performs all data transmission and reception with the other devices via the communication device 83.

[0057] The auxiliary storage device 82 is a computer-readable nonvolatile storage device. The auxiliary storage device 82 stores computer programs executed by the CPU 80 and various data referenced by the CPU 80. For example, one or both of a flash memory and a hard disk drive are used as the auxiliary storage device 82.

[0058] The RAM 81 is a computer-readable volatile storage device that mainly stores the computer programs executed by the CPU 80 and data output and referenced during the execution of the programs by the CPU 80 .

[0059] The CPU 80 includes a plurality of processing modules implemented by executing the computer program. The plurality of processing modules include a main control unit 8a and a job control unit 8b. In addition, some or all of the plurality of processing modules may be implemented by an independent processor such as a DSP that is separate from the CPU 80.

[0060] The main control unit 8a performs processing for selecting a job according to an operation on the operation unit 801 , displaying information on the display unit 802 , and setting various data. The main control unit 8a also performs processing for determining the content of data received by the communication device 83 .

[0061] The job control unit 8b controls the image reading device 1 and the image forming device 2. For example, when the data received by the communication device 83 includes the print data, the job control unit 8b causes the image forming device 2 to execute the print process based on the received data.

[0062] When the main control unit 8a detects a copy request operation on the operation unit 801, the job control unit 8b causes the image reading device 1 to execute the reading process and causes the image forming device 2 to execute the printing process based on the image obtained by the reading process.

[0063] During the printing process, image defects such as vertical lines Ps11, horizontal lines Ps12, noise points Ps13, or density unevenness may occur in the image formed on the output sheet (see Figure 6 、 8 ).

[0064] As described above, the image forming apparatus 2 performs the printing process using an electrophotographic method. In this case, the cause of image defects may be various components, including the photoreceptor 41, the charging device 42, the developing device 43, and the transfer device 44. Furthermore, determining the cause of image defects requires expertise.

[0065] In this embodiment, the image forming apparatus 2 executes a test printing process for forming a predetermined original test image g01 on the sheet.

[0066] For example, when the main control unit 8a detects a test output operation on the operation unit 801, the job control unit 8b causes the image forming apparatus 2 to execute the test print process. In the following description, the sheet on which the original test image g01 is formed is referred to as a test output sheet 9 (see Figure 1 ).

[0067] When executing the test print process, the main control unit 8 a displays a predetermined guidance message on the display unit 802 . The guidance message urges the user to set the test output sheet 9 on the image reading apparatus 1 and then operate the operation unit 801 to start reading.

[0068] Then, when the main control unit 8a detects a reading start operation on the operation unit 801 after displaying the guidance message on the display unit 802, the job control unit 8b causes the image reading device 1 to execute the reading process. Thus, the image reading device 1 reads the original test image g01 from the test output sheet 9 output by the image forming device 2, and obtains a read image corresponding to the original test image g01.

[0069] As will be described later, the CPU 80 executes a process for determining the presence and cause of the image defect based on the read image or the test image g1 which is an image obtained by compressing the read image (see Figure 6 The CPU 80 is an example of a processor that executes an image processing method for determining the presence or absence of the image defect and the cause thereof.

[0070] Alternatively, the device that reads the original test image g01 from the test output sheet 9 may be, for example, a digital camera. The image reading device 1 or the digital camera reading the original test image g01 from the test output sheet 9 is an example of image reading processing on the test output sheet 9 .

[0071] However, the test image g1 may contain multiple types of image defects. In this case, in order to simplify the determination process and improve the determination accuracy, it is preferable to determine the cause of the image defect in the test image for each type of image defect.

[0072] Furthermore, it is preferable that the image processing device 10 can extract the image defective portions by type from the test image g1 through simple processing.

[0073] In the image processing apparatus 10, the CPU 80 executes the image defect determination process described later (see Figure 3 ) Thus, the CPU 80 can extract the image defects occurring in the image forming apparatus 2 by type from the test image g1 through simple processing.

[0074] Furthermore, when determining the cause of the image defect by comparing the value of a specific image parameter such as image color, density, or screen ruling with a predetermined threshold value, omissions in determination or erroneous determinations are likely to occur.

[0075] On the other hand, image pattern recognition processing is suitable for accurately classifying an input image into a large number of phenomena. For example, this processing uses a learning model trained using sample images corresponding to multiple candidate phenomena as training data to determine which of the candidate phenomena an input image corresponds to.

[0076] However, the amount of information in the image is large, and the types of image defects include vertical lines Ps11, horizontal lines Ps12, noise points Ps13, and uneven density (see Figure 6 、 8 ). Furthermore, there are many conceivable candidates for the cause of each type of image failure.

[0077] Therefore, when the test image g1 containing the image defect is used as the input image for the pattern recognition process, the computational complexity of the pattern recognition process becomes extremely large, making it difficult to execute the pattern recognition process using a processor included in a multifunction peripheral or the like.

[0078] Furthermore, to improve the accuracy of determining the cause of image defects, a large amount of training data is required for learning the learning model. However, preparing a large number of test images g1 corresponding to the expected combinations of image defect types and causes for each model of image forming apparatus 2 requires considerable time and effort.

[0079] As will be described later, the CPU 80 performs pattern recognition processing on the image in a manner that can reduce the amount of calculation in the image failure determination process, thereby enabling the cause of the image failure to be determined with high accuracy.

[0080] In the following description, images such as the test image g1 that are processed by the CPU 80 are digital image data. This digital image data constitutes mapping data (map data) that includes multiple pixel values corresponding to two-dimensional coordinate regions in the main scanning direction D1 and the sub-scanning direction D2 that intersects the main scanning direction D1, for each of the three primary colors. The three primary colors are, for example, red, green, and blue. The sub-scanning direction D2 is perpendicular to the main scanning direction D1. Furthermore, the main scanning direction D1 is the horizontal direction of the test image g1, while the sub-scanning direction D2 is the vertical direction of the test image g1.

[0081] For example, the original test image g01 and the test image g1 are mixed color halftone images obtained by synthesizing a plurality of uniform single-color halftone images corresponding to a plurality of developed colors of the image forming device 2. The plurality of single-color halftone images are images uniformly formed at predetermined intermediate grayscale reference densities.

[0082] In this embodiment, the original test image g01 and the test image g1 are mixed-color halftone images obtained by synthesizing four uniform single-color halftone images corresponding to all developed colors of the image forming device 2. During the test print process, a single test output sheet 9 containing the single original test image g01 is output. Therefore, the single test image g1 corresponding to the original test image g01 is the specific target of the image defect.

[0083] Furthermore, the plurality of processing modules of the CPU 80 further include a feature image generating unit 8c, a unique portion determining unit 8d, a color vector determining unit 8e, a periodicity determining unit 8f, a pattern recognizing unit 8g, and a random unevenness determining unit 8h (see FIG. Figure 2 ).

[0084] [Image defect judgment processing]

[0085] Below, refer to Figure 3 The flowchart shown in FIG. 1 illustrates an example of the sequence of the image failure determination process. In the following description, S101, S102, ... represent identification codes for a plurality of steps in the image failure determination process.

[0086] When the reading process is executed according to the reading start operation on the operation unit 801 after the guidance message is displayed on the display unit 802 , the main control unit 8 a causes the characteristic image generating unit 8 c to execute the process of step S101 of the image failure determination process.

[0087] <Step S101>

[0088] In step S101 , the characteristic image generating section 8 c generates a test image g1 from the read image obtained in the image reading process on the test output sheet 9 .

[0089] For example, the characteristic image generating unit 8 c extracts the portion of the original image excluding the blank area at the outer edge from the read image as the test image g1 .

[0090] Alternatively, the characteristic image generation unit 8c generates a test image g1 by performing a compression process that compresses the portion of the original image, excluding the outer edge blank areas, from the read image to a predetermined reference resolution. The characteristic image generation unit 8c compresses the read image when the resolution of the read image is higher than the reference resolution. After generating the test image g1, the main control unit 8a transfers the process to step S102.

[0091] <Step S102>

[0092] In step S102, the characteristic image generating unit 8c starts the specific defect determination process described later. The specific defect determination process is a process for determining the presence or absence of a specific portion Ps1 such as a vertical line Ps11, a horizontal line Ps12, or a noise point Ps13 in the test image g1 and the cause of the specific portion Ps1 (see Figure 6 ) The special portion Ps1 is an example of the image defect.

[0093] When the specific defect determination process is completed, the main control unit 8a shifts the process to step S103.

[0094] <Step S103>

[0095] In step S103, the periodicity determination unit 8f starts a density unevenness determination process described later. When the density unevenness determination process is completed, the main control unit 8a advances the process to step S104.

[0096] <Step S104>

[0097] In step S104 , if the main control unit 8 a determines that the image defect has occurred in the process of step S102 or step S103 , the process proceeds to step S105 ; otherwise, the process proceeds to step S106 .

[0098] <Step S105>

[0099] In step S105 , the main control unit 8 a executes a failure response process that is previously associated with the type and cause of the image failure determined to have occurred in the process of step S102 or step S103 .

[0100] For example, the defect response process includes one or both of the first response process and the second response process described below. The first response process causes the display unit 802 to display a message urging replacement of the component causing the image defect. The second response process corrects image creation parameters to eliminate or alleviate the image defect. The image creation parameters are parameters related to the control of the image creation unit 4x.

[0101] After executing the failure response process, the main control unit 8a ends the image failure determination process.

[0102] <Step S106>

[0103] On the other hand, in step S106 , the main control unit 8 a performs a normal notification indicating that the image failure has not been determined, and then ends the image failure determination process.

[0104] [Specific Defect Judgment and Processing]

[0105] Next, refer to Figure 4 The flowchart shown here illustrates an example of the procedure of the specific defect determination process in step S102. In the following description, S201, S202, ... represent identification codes for a plurality of steps in the specific defect determination process. The specific defect determination process starts with step S201.

[0106] <Step S201>

[0107] First, in step S201, the feature image generator 8c performs a predetermined feature extraction process on the test image g1 to generate a plurality of feature images g21, g22, and g23. Each feature image g21, g22, and g23 is an image obtained by extracting a predetermined specific type of unique portion Ps1 from the test image g1.

[0108] In this embodiment, the plurality of characteristic images g21, g22, and g23 include a first characteristic image g21, a second characteristic image g22, and a third characteristic image g23 (see Figure 6 ).

[0109] The first characteristic image g21 is an image obtained by extracting the vertical line Ps11 in the test image g1. The second characteristic image g22 is an image obtained by extracting the horizontal line Ps12 in the test image g1. The third characteristic image g23 is an image obtained by extracting the noise point Ps13 in the test image g1.

[0110] In this embodiment, the feature extraction process includes a first pre-processing, a second pre-processing, and a unique portion extraction process. In the following description, the pixels sequentially selected from the test image g1 are referred to as focus pixels Px1 (refer to Figure 6 、 7 ).

[0111] The characteristic image generating unit 8c generates a first pre-processed image g11 (see FIG. 11 ) by executing the first pre-processing on the test image g1 with the main scanning direction D1 being the processing direction Dx1. Figure 6 ).

[0112] Furthermore, the characteristic image generating unit 8c generates a second pre-processed image g12 (see FIG. 1 ) by executing the second pre-processing on the test image g1 with the sub-scanning direction D2 as the processing direction Dx1. Figure 6 ).

[0113] Furthermore, the characteristic image generating unit 8 c generates three characteristic images g21 , g22 , and g23 by executing the above-described unique portion extraction process on the first pre-processed image g11 and the second pre-processed image g12 .

[0114] The first pre-processing includes a main filtering process with the main scanning direction D1 as the processing direction Dx1. The main filtering process is a process of transforming the pixel value of the pixel of interest Px1 selected sequentially from the test image g1 into a transformed value obtained by emphasizing the difference between the pixel value of the target area Ax1 and the pixel values of the two adjacent areas Ax2 (see Figure 6 、 7 ).

[0115] The attention area Ax1 is an area including the attention pixel Px1, and the two adjacent areas Ax2 are areas adjacent to the attention area Ax1 on both sides in a predetermined processing direction Dx1. The attention area Ax1 and the adjacent areas Ax2 are areas including one or more pixels each.

[0116] The sizes of the target area Ax1 and the adjacent area Ax2 are set according to the width of the vertical line Ps11 or the horizontal line Ps12 to be extracted, or the size of the noise point Ps13 to be extracted.

[0117] The target area Ax1 and the adjacent area Ax2 each occupy the same range in a direction intersecting the processing direction Dx1. Figure 7 In the example shown, the region of interest Ax1 is a 21-pixel area spanning three columns and seven rows, centered around the pixel of interest Px1. Each adjacent region Ax2 also spans three columns and seven rows, each containing 21 pixels. For each region of interest Ax1 and adjacent region Ax2, the number of rows refers to the number of lines along the processing direction Dx1, while the number of columns refers to the number of lines along a direction intersecting the processing direction Dx1. The size of each region of interest Ax1 and adjacent region Ax2 is predefined.

[0118] In the main filtering process, each pixel value of the target area Ax1 is converted to a first correction value using a predetermined first correction coefficient K1, and each pixel value of each adjacent area Ax2 is converted to a second correction value using a predetermined second correction coefficient K2.

[0119] For example, the first correction coefficient K1 is a coefficient with a digit of 1 or greater that is multiplied by each pixel value of the target area Ax1, and the second correction coefficient K2 is a coefficient less than 0 that is multiplied by each pixel value of the adjacent area Ax2. In this case, the first correction coefficient K1 and the second correction coefficient K2 are set so that the sum of the value obtained by multiplying the number of pixels in the target area Ax1 by the first correction coefficient K1 and the value obtained by multiplying the number of pixels in the two adjacent areas Ax2 by the second correction coefficient K2 is zero.

[0120] The characteristic image generation unit 8c multiplies each pixel value in the target area Ax1 by the first correction coefficient K1 to derive the first correction value corresponding to each pixel in the target area Ax1. It also multiplies each pixel value in the two adjacent areas Ax2 by the second correction coefficient K2 to derive the second correction value corresponding to each pixel in the two adjacent areas Ax2. The characteristic image generation unit 8c then derives a value combining the first and second correction values as the transformed value for the pixel value of the target pixel Px1.

[0121] For example, the characteristic image generating unit 8 c derives the conversion value by adding the total or average value of the first correction values corresponding to the pixels of the target area Ax1 and the total or average value of the second correction values corresponding to the pixels of the two adjacent areas Ax2 .

[0122] The absolute value of the conversion value is a value obtained by magnifying the absolute value of the difference between the pixel value of the target area Ax1 and the pixel values of the two adjacent areas Ax2. The process of deriving the conversion value obtained by combining the first correction value and the second correction value is an example of a process for emphasizing the difference between the pixel value of the target area Ax1 and the pixel values of the two adjacent areas Ax2.

[0123] In addition, a case where the first correction coefficient K1 is a negative number and the second correction coefficient K2 is a positive number may also be considered.

[0124] For example, it is conceivable that the feature image generating unit 8c generates first main map data including a plurality of integrated values obtained by the main filtering process with the main scanning direction D1 as the processing direction Dx1 as the first pre-processing image g11.

[0125] like Figure 6 As shown, when the test image g1 contains one or both of the vertical line Ps11 and the noise point Ps13, the first main mapping data obtained by extracting one or both of the vertical line Ps11 and the noise point Ps13 contained in the test image g1 is generated by the main filtering processing with the main scanning direction D1 as the processing direction Dx1.

[0126] Furthermore, when the test image g1 includes the horizontal line Ps12 , the first main map data is generated by performing the main filtering process with the main scanning direction D1 as the processing direction Dx1 , from which the horizontal line Ps12 included in the test image g1 is removed.

[0127] In addition, the vertical line Ps11 corresponds to the first unique portion, the horizontal line Ps12 corresponds to the second unique portion, and the noise point Ps13 corresponds to the third unique portion.

[0128] On the other hand, the second pre-processing includes the main filtering process in which the sub-scanning direction D2 is used as the processing direction Dx1.

[0129] For example, it is conceivable that the feature image generating unit 8c generates second main map data including a plurality of integrated values obtained by the main filtering process with the sub-scanning direction D2 as the processing direction Dx1 as the second pre-processing image g12.

[0130] like Figure 6 As shown, when the test image g1 contains one or both of the horizontal line Ps12 and the noise point Ps13, the second main mapping data obtained by extracting one or both of the horizontal line Ps12 and the noise point Ps13 contained in the test image g1 is generated by the main filtering processing with the sub-scanning direction D2 as the processing direction Dx1.

[0131] Furthermore, when the test image g1 includes the vertical line Ps11, the second main map data is generated by performing the main filtering process with the sub-scanning direction D2 as the processing direction Dx1, from which the vertical line Ps11 included in the test image g1 is removed.

[0132] However, during the main filtering process, at the edges of the unique portion Ps1 in the processing direction Dx1, an erroneous integrated value may be derived, with the sign being reversed relative to the integrated value representing the original state of the unique portion Ps1. If such an erroneous integrated value is processed as the pixel value representing the unique portion Ps1, it may adversely affect the determination of image defects.

[0133] Therefore, in this embodiment, the first pre-processing includes, in addition to the main filtering process using the main scanning direction D1 as the processing direction Dx1, an edge emphasis filtering process using the main scanning direction D1 as the processing direction Dx1.

[0134] Similarly, the second pre-processing includes, in addition to the main filtering process in which the sub-scanning direction D2 is used as the processing direction Dx1, the edge emphasis filtering process in which the sub-scanning direction D2 is used as the processing direction Dx1.

[0135] The edge emphasis filtering process is a process of performing edge emphasis on a predetermined one of the target area Ax1 and two adjacent areas Ax2.

[0136] Specifically, the edge emphasis filtering process is a process of converting the pixel value of the focus pixel Px1 selected sequentially from the test image g1 into an edge intensity obtained by combining a third correction value obtained by correcting the pixel value of the focus area Ax1 with a positive or negative third correction coefficient K3 and a fourth correction value obtained by correcting the pixel value of the adjacent area Ax2 with a fourth correction coefficient K4 having a positive or negative sign opposite to the third correction coefficient K3 (see FIG. Figure 6 ).

[0137] exist Figure 6 In the example shown, the third correction coefficient K3 is a positive coefficient, and the fourth correction coefficient K4 is a negative coefficient. The third correction coefficient K3 and the fourth correction coefficient K4 are set so that the sum of the value obtained by multiplying the number of pixels in the target area Ax1 by the third correction coefficient K3 and the value obtained by multiplying the number of pixels in the adjacent area Ax2 by the fourth correction coefficient K4 is zero.

[0138] By executing the edge emphasis filter process with the main scanning direction D1 as the processing direction Dx1 , horizontal edge intensity map data is generated in which each pixel value of the test image g1 is converted into the edge intensity.

[0139] Similarly, by executing the edge emphasis filter process with the sub-scanning direction D2 as the processing direction Dx1 , vertical edge intensity map data is generated by converting each pixel value of the test image g1 into the edge intensity.

[0140] In the present embodiment, the characteristic image generating unit 8 c generates the first main map data generated by the main filtering process with the main scanning direction D1 as the processing direction Dx1.

[0141] Furthermore, the characteristic image generating unit 8 c generates the horizontal edge intensity map data by executing the edge emphasis filtering process with the main scanning direction D1 as the processing direction Dx1.

[0142] Furthermore, the feature image generator 8c generates a first pre-processed image g11 by correcting each pixel value of the first main map data using each pixel value of the corresponding horizontal edge intensity map data. For example, the feature image generator 8c generates the first pre-processed image g11 by adding the absolute value of each pixel value of the horizontal edge intensity map data to each pixel value of the first main map data.

[0143] Similarly, the characteristic image generating unit 8 c generates the second main map data by executing the main filter processing with the sub-scanning direction D2 as the processing direction Dx1.

[0144] Furthermore, the characteristic image generating unit 8 c generates the vertical edge intensity map data by executing the edge emphasis filtering process with the sub-scanning direction D2 as the processing direction Dx1 .

[0145] Furthermore, the feature image generator 8c generates a second pre-processed image g12 by correcting each pixel value of the second main map data using each pixel value of the corresponding vertical edge intensity map data. For example, the feature image generator 8c generates the second pre-processed image g12 by adding the absolute value of each pixel value of the vertical edge intensity map data to each pixel value of the second main map data.

[0146] The unique portion extraction process generates three feature images g21, g22, and g23 by separately extracting the vertical lines Ps11, horizontal lines Ps12, and noise points Ps13 contained in the first pre-processed image g11 or the second pre-processed image g12. The three feature images g21, g22, and g23 are the first feature image g21, the second feature image g22, and the third feature image g23.

[0147] The first characteristic image g21 is an image obtained by extracting the characteristic portion Ps1, which is present in the first pre-processed image g11 and is unique to the second pre-processed image g12, from the characteristic portion Ps1, which is composed of one or more significant pixels. The first characteristic image g21 does not include the horizontal line Ps12 or the noise point Ps13, but does include the vertical line Ps11 if it is present in the first pre-processed image g11.

[0148] The significant pixels are pixels that can be distinguished from other pixels by comparing each pixel value of the test image g1 or an index value based on each pixel value with a predetermined threshold value.

[0149] The second characteristic image g22 is an image obtained by extracting, from the characteristic portions Ps1 of the first and second pre-processed images g11 and g12, the characteristic portion Ps1 that exists in the second pre-processed image g12 and is unique to both images. The second characteristic image g22 does not include the vertical line Ps11 or the noise point Ps13, but does include the horizontal line Ps12 if it is present in the second pre-processed image g12.

[0150] The third characteristic image g23 is an image obtained by extracting the common characteristic portion Ps1 in the first pre-processed image g11 and the second pre-processed image g12. The third characteristic image g23 does not contain the vertical line Ps11 or the horizontal line Ps12, but does contain the noise point Ps13 if it is present in the first pre-processed image g11 or the second pre-processed image g12.

[0151] Various methods are conceivable as a method of generating the three characteristic images g21 , g22 , and g23 from the first pre-processed image g11 and the second pre-processed image g12 .

[0152] For example, the characteristic image generating unit 8c derives the index value Zi by applying the first pixel value Xi, which is each pixel value of the first pre-processed image g11 that exceeds a predetermined reference value, and the second pixel value Yi, which is each pixel value of the second pre-processed image g12 that exceeds the reference value, to the following equation (1). Here, the subscript i is an identification number for the position of each pixel.

[0153] [Formula 1]

[0154] Zi=(|Xi|-|Yi|) / (|Xi|+|Yi|)···(1)

[0155] The index value Zi for the pixels forming the vertical line Ps11 is a relatively large positive number. Meanwhile, the index value Zi for the pixels forming the horizontal line Ps12 is a relatively small negative number. Furthermore, the index value Zi for the pixels forming the noise point Ps13 is 0 or close to 0. The index value Zi is an example of an index value representing the difference between the corresponding pixel values in the first pre-processed image g11 and the second pre-processed image g12.

[0156] The above-mentioned properties of the index value Zi can be used to simplify the process of extracting vertical lines Ps11 from the first pre-processed image g11, horizontal lines Ps12 from the second pre-processed image g12, and noise points Ps13 from the first pre-processed image g11 or the second pre-processed image g12.

[0157] For example, the characteristic image generating unit 8c generates a first characteristic image g21 by transforming the first pixel value Xi of the first pre-processed image g11 to the first specificity Pi derived by the following equation (2). Thus, the first characteristic image g21 is generated by extracting the vertical line Ps11 from the first pre-processed image g11.

[0158] [Formula 2]

[0159] Pi=Xi·Zi···(2)

[0160] Furthermore, the characteristic image generating unit 8c converts the second pixel value Yi of the second pre-processed image g12 to the second specificity Qi derived by the following equation (3) to generate a second characteristic image g22. Thus, the second characteristic image g22 is generated by extracting the horizontal line Ps12 from the second pre-processed image g12.

[0161] [Formula 3]

[0162] Qi=Yi·(-Zi)···(3)

[0163] Furthermore, the characteristic image generating unit 8c generates a third characteristic image g23 by transforming the first pixel value Xi of the first pre-processed image g11 to the third specificity Ri derived by the following equation (4). Thus, the third characteristic image g23 is generated by extracting the noise point Ps13 from the first pre-processed image g11.

[0164] [Formula 4]

[0165] Ri=Xi·(1-Xi)…(4)

[0166] Alternatively, the characteristic image generating unit 8c may generate a third characteristic image g23 by transforming the second pixel value Yi of the second pre-processed image g12 to the third specificity Ri derived by the following equation (5). In this way, the third characteristic image g23 is generated by extracting the noise point Ps13 from the second pre-processed image g12.

[0167] [Formula 5]

[0168] Ri=Yi·(Zi-1)…(5)

[0169] As described above, the characteristic image generating unit 8c generates the first characteristic image g21 by transforming each pixel value of the first pre-processed image g11 using the predetermined formula (2) based on the index value Zi. Formula (2) is an example of a first transformation formula.

[0170] Furthermore, the characteristic image generating unit 8c generates the second characteristic image g22 by transforming each pixel value of the second pre-processed image g12 using the predetermined formula (3) based on the index value Zi. Formula (3) is an example of the second transformation formula.

[0171] Furthermore, the characteristic image generating unit 8c generates a third characteristic image g23 by transforming each pixel value of the first pre-processed image g11 or the second pre-processed image g12 using the predetermined equation (4) or (5) based on the index value Zi. Equations (4) and (5) are examples of third transformation equations.

[0172] The process of generating the first characteristic image g21, the second characteristic image g22 and the third characteristic image g23 in step S201 is an example of processing of extracting vertical lines Ps11, horizontal lines Ps12 and noise points Ps13 from one or more special parts Ps1 of the first pre-processed image g11 and the second pre-processed image g12 as the aforementioned image defects.

[0173] After generating the characteristic images g21 , g22 , and g23 , the characteristic image generating unit 8 c shifts the process to step S202 .

[0174] <Step S202>

[0175] In step S202 , the unique portion identifying unit 8 d identifies the position of the unique portion Ps1 of each of the characteristic images g21 , g22 , and g23 .

[0176] For example, the unique portion identifying unit 8 d determines that a portion of each of the characteristic images g21 , g22 , and g23 having a pixel value deviating from a predetermined reference range is a unique portion Ps1 .

[0177] In addition, for each characteristic image g21, g22, and g23, when multiple specific parts Ps1 exist within a predetermined proximity range in each main scanning direction D1 and sub-scanning direction D2, the specific part determination unit 8d performs a combination process of combining these multiple specific parts Ps1 into a series of one specific part Ps1.

[0178] For example, when the first characteristic image g21 includes two vertical lines Ps11 spaced apart in the sub-scanning direction D2 within the proximity range, the unique portion identifying unit 8 d combines the two vertical lines Ps11 into one vertical line Ps11 through the combining process.

[0179] Similarly, when the second characteristic image g22 includes two horizontal lines Ps12 spaced apart in the main scanning direction D1 within the proximity range, the unique portion identifying unit 8 d combines these two horizontal lines Ps12 into one horizontal line Ps12 through the combining process.

[0180] In addition, when the third characteristic image g23 includes multiple noise points Ps13 arranged at intervals in the main scanning direction D1 or the sub-scanning direction D2 within the said proximity range, the special part determination unit 8d combines these multiple noise points Ps13 into one noise point Ps13 through the said combination processing.

[0181] If the position of the unique portion Ps1 is not determined in any of the three characteristic images g21, g22, and g23, the unique portion identifying unit 8d terminates the above-described specific defect determination process. On the other hand, if the position of the unique portion Ps1 is determined in one or more of the three characteristic images g21, g22, and g23, the unique portion identifying unit 8d proceeds to step S203.

[0182] <Step S203>

[0183] In step S203 , the color vector determination unit 8 e specifies a color vector representing a vector in a color space extending from the color of the unique portion Ps1 of the test image g1 and the color of the reference region including the surroundings of the unique portion Ps1 to the other direction.

[0184] The reference region is a predetermined region defined using the unique portion Ps1 as a reference. For example, the reference region includes the surrounding region adjacent to the unique portion Ps1 but does not include the unique portion Ps1. Alternatively, the reference region may include both the unique portion Ps1 and the surrounding region adjacent to the unique portion Ps1.

[0185] The test image g1 is originally a uniform halftone image. Therefore, when a good test image g1 is formed on the test output sheet 9, the unique portion Ps1 cannot be identified, and the color vector at any position in the test image g1 is substantially a zero vector.

[0186] On the other hand, when the unique portion Ps1 is identified, the direction of the color vector between the unique portion Ps1 and the reference area corresponding to the unique portion Ps1 indicates excess or deficiency in toner density of any one of the four developing colors of the image forming apparatus 2 .

[0187] Therefore, the direction of the color vector indicates which of the four image generating units 4 x of the image forming apparatus 2 is responsible for the occurrence of the unique portion Ps1 .

[0188] Alternatively, the color vector specifying unit 8e may specify a vector in the color space extending from one of the color of the unique portion Ps1 in the test image g1 and a predetermined reference color to the other as the color vector. In this case, the reference color is the original color of the test image g1.

[0189] Furthermore, in step S203, the color vector determination unit 8e determines the developed color that causes the unique portion Ps1 and the excess or deficiency state of the developed color density based on the color vector.

[0190] For example, the auxiliary storage device 82 pre-stores information on a plurality of unit vectors indicating directions in which the density of cyan, magenta, yellow, or black increases and directions in which the density decreases relative to the reference color of the test image g1.

[0191] The color vector determination unit 8e normalizes the color vector to a predetermined unit length. Furthermore, the color vector determination unit 8e determines which of the plurality of unit vectors corresponding to the increased or decreased density of cyan, magenta, yellow, or black is most similar to the normalized color vector, thereby determining the developing color that is the cause of the unique portion Ps1 and the excess or deficiency of the developing color's density.

[0192] Then, after executing the process of step S203 , the color vector determination unit 8 e shifts the process to step S204 .

[0193] <Step S204>

[0194] In step S204, when the specific portion Ps1 is determined in one or both of the second characteristic image g22 and the third characteristic image g23, the periodicity determination unit 8f transfers the processing to step S205. When the specific portion Ps1 is not determined, the periodicity determination unit 8f transfers the processing to step S206.

[0195] In the following description, one or both of the second characteristic image g22 and the third characteristic image g23 in which the unique portion Ps1 is determined are referred to as periodic determination target images. The unique portion Ps1 of the periodic determination target image is the horizontal line Ps12 or the noise point Ps13 (see Figure 6 ).

[0196] <Step S205>

[0197] In step S205, the periodicity determination unit 8f performs a periodic unique portion determination process on the periodicity determination target image. The periodic unique portion determination process includes a quantity determination process, a unique portion periodicity determination process, and a unique portion periodicity cause determination process.

[0198] The number determination process is a process of determining the number of unique portions Ps1 arranged in the sub-scanning direction D2 in the periodicity determination target image.

[0199] Specifically, the periodicity determination unit 8f determines the number of horizontal lines Ps12 arranged in the sub-scanning direction D2 by counting the number of horizontal lines Ps12 arranged in the sub-scanning direction D2 that occupy the same range in the main scanning direction D1 in the second characteristic image g22 and exceed a predetermined ratio.

[0200] Furthermore, the periodicity determination unit 8f determines the number of noise points Ps13 arranged in the sub-scanning direction D2 by counting the number of noise points Ps13 whose position deviation in the main scanning direction D1 in the third characteristic image g23 is within a predetermined range and arranged in the sub-scanning direction D2.

[0201] The periodicity determination unit 8 f executes the unique portion periodicity determination process only for two or more unique portions Ps1 arranged in the sub-scanning direction D2 .

[0202] Furthermore, the periodicity determination unit 8f determines that there is no periodicity for one unique portion Ps1 arranged in the sub-scanning direction D2, and skips the unique portion periodicity determination process and the unique portion periodicity cause determination process.

[0203] The unique portion periodicity determination process is a process for determining the presence or absence of one or more predetermined periodicities in the sub-scanning direction D2 for the periodicity determination target image.

[0204] This periodicity corresponds to the outer circumference of a rotating body associated with image formation, such as the photoreceptor 41, charging roller 42a, developing roller 43a, or primary transfer roller 441, in each image forming unit 4x or transfer device 44. The state of these rotating bodies associated with image formation affects the quality of the image formed on the sheet. In the following description, these rotating bodies associated with image formation are referred to as image formation rotating bodies.

[0205] When the image defect occurs due to a defect in the image forming rotating body, the periodicity corresponding to the outer circumference of the image forming rotating body may appear as intervals in the sub-scanning direction D2 between multiple horizontal lines Ps12 or multiple noise points Ps13.

[0206] Therefore, when the periodic judgment object image has the periodicity corresponding to the outer circumferential length of the image production rotating body, it can be said that the image production rotating body corresponding to the periodicity is the cause of the horizontal line Ps12 or noise point Ps13 in the periodic judgment object image.

[0207] When the number of the unique portions Ps1 aligned in the sub-scanning direction D2 in the periodicity determination target image is two, the periodicity determination unit 8 f performs an interval derivation process as the unique portion periodicity determination process.

[0208] In the interval derivation process, the periodicity determination unit 8 f derives the interval between the two unique portions Ps1 in the sub-scanning direction D2 as the period of the two unique portions Ps1 .

[0209] When the number of the unique portions Ps1 arranged in the sub-scanning direction D2 in the periodicity determination target image is three or more, the periodicity determination unit 8 f executes a frequency analysis process as the unique portion periodicity determination process.

[0210] The periodicity judgment unit 8f performs frequency analysis such as Fourier transform on the periodicity judgment object image containing three or more specific parts Ps1 arranged in the sub-scanning direction D2 in the frequency analysis processing, thereby determining the dominant frequency in the frequency distribution of the data string of the specific part Ps1 in the periodicity judgment object image, that is, the specific part frequency.

[0211] Furthermore, the periodicity determination unit 8f derives a period corresponding to the frequency of the unique portion as the period of the three or more unique portions Ps1.

[0212] Furthermore, during the unique portion periodicity cause determination process, the periodicity determination unit 8f determines whether the outer circumference of each of the plurality of predetermined candidates for the image-forming rotating body satisfies a predetermined periodicity approximation condition relative to the period of the unique portion Ps1. The plurality of candidates for the image-forming rotating body in step S205 are examples of the plurality of predetermined cause candidates corresponding to the horizontal line Ps12 or the noise point Ps13.

[0213] In the following description, among the special parts Ps1 contained in the second characteristic image g22 and the third characteristic image g23, the special part Ps1 of an object that is judged to be a candidate for the image production rotating body and satisfies the periodic approximation condition will be called a periodic special part, and the other special parts Ps1 will be called non-periodic special parts.

[0214] The periodicity determination unit 8f determines that one of the candidate image creation rotating bodies that satisfies the periodic approximation condition is the cause of the periodic unique portion in the unique portion periodicity cause determination process, thereby determining the cause of the horizontal line Ps12 or the noise point Ps13.

[0215] In step S205, the periodicity determination unit 8f determines, based on the color vector determined in step S203, which of the four image creation units 4x having different development colors is responsible for the horizontal line Ps12 or the noise point Ps13.

[0216] When three or more unique portions Ps1 arranged in the sub-scanning direction D2 include the aperiodic unique portion that does not correspond to the unique portion frequency, the periodicity determination unit 8f subjects the aperiodic unique portion to characteristic pattern recognition processing described later.

[0217] For example, the periodicity determination unit 8f generates inverse Fourier transform data by performing inverse Fourier transform on the frequency distribution obtained by the Fourier transform, from which frequency components other than the unique portion frequency are removed.

[0218] Furthermore, the periodicity determination unit 8f determines that a characteristic portion, among the three or more characteristic portions Ps1 arranged in the sub-scanning direction D2, that is located at a position deviated from the peak position of the waveform in the sub-scanning direction D2 represented by the inverse Fourier transform data is the aperiodic characteristic portion.

[0219] Then, when it is determined as a result of the process in step S205 that the second characteristic image g22 and the third characteristic image g23 do not include the aperiodic specific portion, the periodicity determination unit 8 f ends the specific failure determination process.

[0220] On the other hand, when it is determined as a result of the process of step S205 that the second characteristic image g22 and the third characteristic image g23 include the aperiodic specific portion, the process proceeds to step S206 .

[0221] <Step S206>

[0222] In step S206, the pattern recognition unit 8g performs feature pattern recognition processing on the first feature image g21 and the second and third feature images g22 and g23, each of which includes the aperiodic feature portion. The second feature image g22 and the third feature image g23, each of which includes the aperiodic feature portion, are examples of aperiodic feature images.

[0223] In the characteristic pattern recognition process, the first characteristic image g21, and the second and third characteristic images g22 and g23, each including the aperiodic characteristic portion, are used as input images. In the characteristic pattern recognition process, the pattern recognition unit 8g determines, through pattern recognition of the input image, which of a plurality of predetermined candidate causes of the image defect the input image corresponds to.

[0224] Furthermore, the input image for the feature pattern recognition process may also include the horizontal edge intensity map data or the vertical edge intensity map data obtained by the edge emphasis filtering process. For example, in the feature pattern recognition process for determining the vertical line Ps11, the first feature image g21 and the horizontal edge intensity map data are used as the input image.

[0225] Likewise, in the feature pattern recognition process for determining the horizontal line Ps12 , the second feature image g22 and the vertical edge intensity map data are used as the input images.

[0226] Likewise, in the feature pattern recognition process for determining the noise point Ps13 , the third feature image g23 and one or both of the horizontal edge intensity map data and the vertical edge intensity map data are used as the input image.

[0227] For example, the feature pattern recognition process is a process of classifying the input image into any one of the plurality of cause candidates using a learning model obtained by previously learning a plurality of sample images corresponding to the plurality of cause candidates as training data.

[0228] For example, the learning model is a model that uses a classification machine learning algorithm called random forest, a model that uses a machine learning algorithm called SVM (Support Vector Machine), or a model that uses a CNN (Convolutional Neural Network) algorithm.

[0229] The learning model is prepared separately for the first characteristic image g21, and the second characteristic image g22 and the third characteristic image g23 each including the aperiodic characteristic portion. In addition, the plurality of sample images are used as the training data for each of the cause candidates.

[0230] In step S206, the pattern recognition unit 8g determines which component of the four image creation units 4x having different development colors is the cause of the vertical line Ps11, horizontal line Ps12, or noise point Ps13 based on the color vector determined in step S203.

[0231] The process of step S206 determines the cause of the vertical line Ps11 and the causes of the horizontal line Ps12 and noise point Ps13 identified as the aperiodic characteristic portion. After executing the process of step S206, the pattern recognition unit 8g ends the characteristic failure determination process.

[0232] [Density unevenness judgment processing]

[0233] Next, refer to Figure 5 The flowchart shown in FIG. 1 illustrates an example of the procedure of the density unevenness determination process in step S103. In the following description, S301, S302, ... represent identification codes for a plurality of steps in the density unevenness determination process. The density unevenness determination process starts at step S301.

[0234] <Step S301>

[0235] In step S301, the periodicity determination unit 8f derives a vertical data string VD1 for each predetermined specific color of the test image g1. The specific color is a color corresponding to the developed color of the image forming device 2. The vertical data string VD1 is a data string of representative values V1 of a plurality of pixel values in each row in the main scanning direction D1 of the image of the specific color constituting the test image g1 (see Figure 7 ).

[0236] For example, the specific colors are three of the four developing colors of the image forming apparatus 2. In this case, the periodicity determination unit 8f converts the red, green, and blue image data constituting the test image g1 into cyan, yellow, and magenta image data.

[0237] Then, the periodicity determination unit 8f derives representative values V1 of multiple pixel values in each row of the three specific color image data corresponding to the test image g1, and derives three vertical data strings VD1 corresponding to cyan, yellow and magenta.

[0238] Alternatively, the specific colors may be the three primary colors of red, green, and blue. In this case, the periodicity determination unit 8f converts each pixel value of the three image data items of red, green, and blue in the test image g1 into a value representing a ratio to the average or total value of each pixel value of the three image data items of red, green, and blue in the test image g1. Furthermore, the periodicity determination unit 8f derives three vertical data strings VD1 for the three converted image data items.

[0239] Here, red corresponds to cyan, green corresponds to magenta, and blue corresponds to yellow. Specifically, cyan density unevenness appears as density unevenness in the converted red image data, magenta density unevenness appears as density unevenness in the converted green image data, and yellow density unevenness appears as density unevenness in the converted blue image data.

[0240] For example, the representative value V1 is the average value, maximum value, or minimum value of the remaining pixel values after removing the pixel values of the unique portion Ps1 from all pixel values in the row in the main scanning direction D1. Alternatively, the representative value V1 may be the average value, maximum value, or minimum value of all pixel values in the row in the main scanning direction D1.

[0241] Then, after executing the process of step S301 , the periodicity determination unit 8 f shifts the process to step S302 .

[0242] <Step S302>

[0243] In step S302 , the periodicity determination unit 8 f executes periodicity unevenness determination processing for each vertical data string VD1 of the specific color.

[0244] For example, the periodicity determination unit 8f performs frequency analysis such as Fourier transform on each vertical data string VD1 to thereby determine the dominant frequency of the frequency distribution of the vertical data string VD1, that is, the density unevenness frequency.

[0245] Furthermore, the periodicity determination unit 8 f derives a period corresponding to the density unevenness frequency as the period of the density unevenness of the test image g1 .

[0246] Furthermore, the periodicity determination unit 8f determines whether the outer perimeter of each of the plurality of predetermined candidates for the image forming rotating body satisfies the periodic approximation condition with respect to the period of the density unevenness. Determination that any of the plurality of candidates for the image forming rotating body satisfies the periodic approximation condition means that periodic density unevenness has been determined to have occurred in the test image g1.

[0247] The plurality of candidates for the image creation rotating body in step S302 are an example of a plurality of predetermined cause candidates corresponding to the periodic density unevenness in the test image g1. The periodic density unevenness is an example of image failure.

[0248] Furthermore, the periodicity determination unit 8f determines the cause of the periodic density unevenness based on the developed color corresponding to the vertical data string VD1 and the candidate of the image forming rotating body determined to satisfy the periodic approximation condition.

[0249] However, when the periodic density unevenness is caused by the black image forming portion 4 x , variations in pixel values occur in all of the red, green, and blue image data constituting the test image g1 .

[0250] Therefore, when the periodic density unevenness having the periodicity common to all of cyan, magenta, and yellow is determined to occur, the periodicity determination unit 8f determines that the black image creation unit 4x is the cause of the periodic density unevenness.

[0251] Then, when the periodicity determination unit 8 f determines that the periodic density unevenness occurs in the test image g1 , it ends the density unevenness determination process. Otherwise, it shifts the process to step S303 .

[0252] <Step S303>

[0253] In step S303 , the random unevenness determination unit 8 h determines whether random density unevenness, which is a type of image defect, occurs in each of the three specific color image data corresponding to the test image g1 .

[0254] The random density unevenness determination unit 8h determines whether or not the deviation of pixel values of each of the three specific color image data exceeds a predetermined allowable range, thereby determining whether or not the random density unevenness occurs.

[0255] For example, the magnitude of the deviation of the pixel values is determined based on the variance, standard deviation, or the difference between the central value and the maximum value and the minimum value of each of the three specific color image data.

[0256] However, when the random density unevenness determination unit 8h determines that the random density unevenness occurs in all of cyan, magenta, and yellow, it determines that the black image creation unit 4x is the cause of the random density unevenness.

[0257] Then, when the random unevenness determination unit 8h determines that the random density unevenness occurs in the test image g1, the process proceeds to step S304. Otherwise, the random unevenness determination process ends.

[0258] <Step S304>

[0259] In step S304, the pattern recognition unit 8g performs random pattern recognition processing. The random pattern recognition processing is a process that uses the test image g1 determined to have the random density unevenness as an input image and determines which of the one or more cause candidates the input image corresponds to through pattern recognition of the input image.

[0260] Then, after executing the process of step S304 , the pattern recognition unit 8 g ends the density unevenness determination process.

[0261] The image failure determination process including the specific failure determination process and the density unevenness determination process executed by the CPU 80 is an example of an image processing method for determining the cause of the image failure based on the test image g1 read from the output sheet of the image forming apparatus 2 .

[0262] As described above, the feature image generating unit 8c performs the first pre-processing including the main filtering process with the horizontal direction of the test image g1 as the processing direction Dx1, thereby generating a first pre-processed image g11. The main filtering process is a process of transforming the pixel values of the target pixels Px1 selected sequentially from the test image g1 to the transformed values obtained by emphasizing the difference between the pixel values of the target area Ax1 and the pixel values of the two adjacent areas Ax2 adjacent to the target area Ax1 on both sides of the predetermined processing direction Dx1 (see Figure 4 Step S201 and Figure 6 ).

[0263] Furthermore, the feature image generating unit 8c performs the second pre-processing including the main filtering process with the longitudinal direction of the test image g1 as the processing direction Dx1, thereby generating a second pre-processed image g12 (see Figure 4 Step S201 and Figure 6 ).

[0264] Furthermore, the characteristic image generating unit 8c extracts vertical lines Ps11, horizontal lines Ps12 and noise points Ps13 from one or more characteristic parts Ps1 of the first pre-processed image g11 and the second pre-processed image g12 as the image defects (see Figure 4 Step S201 and Figure 6 ).

[0265] The feature extraction process of step S201 is a simple process with a low computational load. By this simple process, three feature images g21, g22, and g23 can be generated by extracting the unique portion Ps1 having different shapes from one test image g1.

[0266] Then, the periodicity determination unit 8f and the pattern recognition unit 8g perform the periodicity characteristic portion determination process and the characteristic pattern recognition process using the first characteristic image g21, the second characteristic image g22, and the third characteristic image g23 to determine the causes of the vertical line Ps11, the horizontal line Ps12, and the noise point Ps13, which are respectively one type of image defects (see Figure 4 Step S205 and step S206, Figure 6 ).

[0267] By individually determining the cause of the image failure for the three characteristic images g21 , g22 , and g23 including the unique portion Ps1 of different types, the cause of the image failure can be determined with high accuracy through a relatively simple determination process.

[0268] The periodic characteristic portion judgment processing of step S205 is as follows: judging the presence or absence of one or more predetermined periodicities in the sub-scanning direction D2 for the second characteristic image g22 or the third characteristic image g23, and judging the cause of the horizontal line Ps12 or the noise point Ps13 based on the judgment result of the periodicity.

[0269] In the case where the horizontal line Ps12 or the noise point Ps13 is caused by a defect in the rotating body associated with the image production, the cause of the horizontal line Ps12 or the noise point Ps13 can be determined with high precision by judging the periodic characteristic portion judgment processing corresponding to the outer circumferential length of the rotating body.

[0270] In addition, the feature pattern recognition processing of step S206 is a processing as follows: through pattern recognition of the input image, it is determined that the input image corresponds to which of the predetermined multiple cause candidates corresponding to the vertical line, the horizontal line, and the noise point. Here, the feature image that is determined to have no periodicity by the periodic characteristic portion determination processing among the first feature image g21, the second feature image g22, and the third feature image g23 is the input image of step S206 (refer to Figure 4 Steps S204 to S206).

[0271] The periodic unique portion determination process of step S205 and the characteristic pattern recognition process of step S206 are examples of a predetermined cause determination process using the first characteristic image g21 , the second characteristic image g22 , and the third characteristic image g23 .

[0272] The characteristic pattern recognition processing using a learning model or the like is performed on each of the characteristic images g21, g22, and g23 obtained by extracting a specific type of unique portion Ps1. This reduces the amount of computation required by the CPU 80 and enables highly accurate determination of the cause of the image defect. Furthermore, by preparing only a relatively small amount of training data corresponding to each specific type of unique portion Ps1, the learning model for each type of unique portion Ps1 can be fully learned.

[0273] In addition, the characteristic pattern recognition process of step S206 is performed on the first characteristic image g21 that is not the subject of the periodic characteristic portion determination process of step S205, and the second characteristic image g22 or the third characteristic image g23 that is determined to have no periodicity by the periodic characteristic portion determination process of step S205 (see Figure 4 Steps S204 to S206).

[0274] In this case, the feature pattern recognition process can exclude the possibility that the image defect is caused by the periodicity of the rotating body associated with the image generation, thereby further simplifying the feature pattern recognition process.

[0275] The color vector determination unit 8e determines the color vector that represents a vector in the color space from the color of the unique portion Ps1 in the test image g1 to the color of the reference area including the surroundings of the unique portion Ps1 (see FIG. Figure 4 Step S203).

[0276] Furthermore, the periodicity determination unit 8f in step S205 and the pattern recognition unit 8g in step S206 use the color vectors in the cause determination process to determine the causes of the vertical line Ps11, the horizontal line Ps12, and the noise point Ps13. Specifically, the periodicity determination unit 8f and the pattern recognition unit 8g use the color vectors to determine which of the multiple developing colors of the image forming apparatus 2 the cause of the image defect corresponds to.

[0277] In the image forming apparatus 2 capable of printing color images, by using the color vector, it is possible to easily and reliably determine which color among a plurality of developed colors the image defect is caused by.

[0278] In addition, the periodicity determination unit 8f performs the periodicity unevenness determination process for each of the predetermined specific colors on the test image g1 (see Figure 5 The periodic unevenness determination process is a process that determines the presence or absence of one or more predetermined periodicities in the sub-scanning direction D2, and further determines the presence or absence of the periodic density unevenness as a type of image defect and the cause thereof based on the periodicity determination result.

[0279] The periodic unevenness determination process makes it possible to determine the cause of the periodic density unevenness with high accuracy.

[0280] Furthermore, the random unevenness determination unit 8h determines whether the deviation of the pixel value exceeds a predetermined allowable range for each specific color in the test image g1 determined to have no periodicity by the periodic unevenness determination process in step S302, thereby determining whether the random density unevenness occurs (see Figure 5 The random density unevenness is a type of image defect.

[0281] Furthermore, the pattern recognition unit 8g performs the random pattern recognition process using the test image g1 determined to have the random density unevenness as the input image (see Figure 5 In the random pattern recognition process, it is determined by pattern recognition of the input image which of the one or more cause candidates the input image corresponds to.

[0282] Furthermore, the test image g1 is a mixed color halftone image obtained by synthesizing a plurality of uniform single-color halftone images corresponding to a plurality of developed colors of the image forming device 2. Thus, the CPU 80 can determine the cause of the image defect for all developed colors of the image forming device 2 using the test image g1 having fewer developed colors than the number used by the image forming device 2.

[0283] [First application example]

[0284] Next, refer to Figure 9 The flowchart shown here explains the procedure of the characteristic image generation process in the first application example of the image processing device 10 .

[0285] In the following description, S401, S402, ... represent identification symbols of a plurality of steps in the feature image generation process of this application example. The feature image generation process of this application example starts from step S401.

[0286] <Step S401>

[0287] In step S401 , the characteristic image generating unit 8 c selects a compression rate to be adopted from a plurality of compression rate candidates set in advance, and the process proceeds to step S402 .

[0288] <Step S402>

[0289] In step S402, the characteristic image generating unit 8c compresses the read image using the selected compression ratio to generate a test image g1. The processing in steps S401 and S402 is an example of compression processing. Thereafter, the characteristic image generating unit 8c transfers the processing to step S403.

[0290] <Step S403>

[0291] In step S403, the characteristic image generating unit 8c performs the first pre-processing on the compressed test image g1 obtained in step S402, thereby generating a first pre-processed image g11. Thereafter, the characteristic image generating unit 8c shifts the process to step S404.

[0292] <Step S404>

[0293] In step S404, the characteristic image generating unit 8c performs the second pre-processing on the compressed test image g1 obtained in step S402, thereby generating a second pre-processed image g12. Thereafter, the characteristic image generating unit 8c shifts the process to step S405.

[0294] <Step S405>

[0295] In step S405 , if the processing of steps S401 to S404 has been executed for all of the plurality of compression ratio candidates, the characteristic image generating unit 8 c proceeds to step S406 . Otherwise, the processing of steps S401 to S404 is executed for different compression ratios.

[0296] In the compression processing of steps S401 and S402 , the characteristic image generating unit 8 c compresses the read image at a plurality of compression ratios, thereby generating a plurality of test images g1 having different sizes.

[0297] Furthermore, in steps S403 and S404 , the feature image generating unit 8 c performs the first preprocessing and the second preprocessing on the test images g1 , thereby generating first preprocessed images g11 and second preprocessed images g12 corresponding to the test images g1 .

[0298] <Step S406>

[0299] In step S406, the characteristic image generation unit 8c performs the aforementioned unique portion extraction process on each of the plurality of first pre-processed images g11 and the plurality of second pre-processed images g12. This process generates a plurality of candidates for each of the first characteristic image g21, the second characteristic image g22, and the third characteristic image g23 corresponding to the plurality of test images g1. The characteristic image generation unit 8c then proceeds to step S407.

[0300] <Step S407>

[0301] In step S407, the characteristic image generating unit 8c generates a first characteristic image g21, a second characteristic image g22, and a third characteristic image g23 by combining the plurality of candidates obtained in step S406. Thereafter, the characteristic image generating unit 8c ends the characteristic image generating process.

[0302] For example, the characteristic image generating unit 8c sets a representative value such as the maximum value or average value of each pixel value among the plurality of candidates for the first characteristic image g21 as each pixel value of the first characteristic image g21. The same applies to the second characteristic image g22 and the third characteristic image g23.

[0303] The processing of steps S401 to S404 is an example of generating a plurality of first pre-processed images g11 and a plurality of second pre-processed images g12 by performing the first pre-processing and the second pre-processing a plurality of times with different size ratios between the test image g1 and the region of interest Ax1 and the adjacent region Ax2. Changing the compression rate is an example of changing the size ratio between the test image g1 and the region of interest Ax1 and the adjacent region Ax2.

[0304] The processing of steps S406 to S407 is an example of a process of generating the first characteristic image g21 , the second characteristic image g22 , and the third characteristic image g23 by the above-described unique portion extraction process based on the plurality of first pre-processed images g11 and the plurality of second pre-processed images g12 .

[0305] By adopting this application example, vertical lines Ps11 or horizontal lines Ps12 of different thicknesses, or noise points Ps13 of different sizes can be extracted without omission.

[0306] [Second application example]

[0307] Next, refer to Figure 10 The flowchart shown here explains the procedure of the characteristic image generation process in the second application example of the image processing device 10 .

[0308] In the following description, S501, S502, ... represent identification symbols of a plurality of steps in the feature image generation process of this application example. The feature image generation process of this application example starts from step S501.

[0309] <Steps S501 to S505>

[0310] The characteristic image generating unit 8c performs the same processing as steps S401 to S405 in steps S501 to S505. In step S505, if the characteristic image generating unit 8c has performed the processing of steps S501 to S504 on all of the plurality of compression ratio candidates, the processing proceeds to step S506.

[0311] <Step S506>

[0312] In step S506, the characteristic image generating unit 8c combines the plurality of first pre-processed images g11 and the plurality of second pre-processed images into one, and then proceeds to step S507.

[0313] For example, the characteristic image generating unit 8c sets a representative value such as the maximum value or average value of each pixel value in the plurality of first pre-processed images g11 as each pixel value of the aggregated first characteristic image g21. The same applies to the plurality of second pre-processed images g12.

[0314] <Step S506>

[0315] In step S506, the characteristic image generation unit 8c performs the above-described unique portion extraction process on the combined first pre-processed image g11 and second pre-processed image g12, thereby generating a first characteristic image g21, a second characteristic image g22, and a third characteristic image g23. The characteristic image generation unit 8c then terminates the characteristic image generation process.

[0316] When this application example is adopted, the same effects as those of the first application example can be obtained.

[0317] [Third Application Example]

[0318] Next, refer to Figure 11 The flowchart shown here explains the procedure of the characteristic image generation process in the third application example of the image processing device 10 .

[0319] In the following description, S601, S602, ... represent identification symbols of a plurality of steps in the feature image generation process of this application example. The feature image generation process of this application example starts from step S601.

[0320] In the following description, the sizes of the target area Ax1 and the adjacent area Ax2 in the first pre-processing and the second pre-processing are referred to as filter sizes.

[0321] <Step S601>

[0322] In step S601 , the characteristic image generating unit 8 c selects the filter size to be adopted from a plurality of size candidates set in advance, and the process proceeds to step S602 .

[0323] <Step S602>

[0324] In step S602, the characteristic image generating unit 8c performs the first pre-processing on the test image g1 using the filter size selected in step S601, thereby generating a first pre-processed image g11. Thereafter, the characteristic image generating unit 8c shifts the process to step S603.

[0325] <Step S603>

[0326] In step S603, the characteristic image generation unit 8c performs the second pre-processing on the test image g1 using the filter size selected in step S601, thereby generating a second pre-processed image g12. Thereafter, the characteristic image generation unit 8c shifts the process to step S604.

[0327] <Step S604>

[0328] In step S604 , if the processing of steps S601 to S603 has been executed for all of the plurality of size candidates, the feature image generating unit 8 c proceeds to step S605 . Otherwise, the processing of steps S601 to S603 is executed using different filter sizes.

[0329] In steps S601 to S604, the feature image generator 8c performs multiple first pre-processing and multiple second pre-processing on a single test image g1, each with a different size of the target area Ax1 and the adjacent area Ax2. This generates multiple first pre-processed images g11 and multiple second pre-processed images g12.

[0330] <Steps S605, S606>

[0331] The feature image generating unit 8c performs the same operation as in steps S605 and S606. Figure 9 The same processing as steps S406 and S407 is performed. Thereafter, the feature image generating unit 8c ends the feature image generating processing.

[0332] Through the processing of steps S605 and S606 , multiple candidates of the first characteristic image g21 , the second characteristic image g22 , and the third characteristic image g23 are aggregated to generate the aggregated first characteristic image g21 , the second characteristic image g22 , and the third characteristic image g23 .

[0333] The processing of steps S601 to S604 is an example of generating a plurality of first pre-processed images g11 and a plurality of second pre-processed images g12 by performing the first pre-processing and the second pre-processing a plurality of times with different size ratios between the test image g1 and the region of interest Ax1 and the adjacent region Ax2. Changing the filter size is an example of changing the size ratio between the test image g1 and the region of interest Ax1 and the adjacent region Ax2.

[0334] By adopting this application example, vertical lines Ps11 or horizontal lines Ps12 of different thicknesses, or noise points Ps13 of different sizes can be extracted without omission.

[0335] [Fourth Application Example]

[0336] Next, refer to Figure 11 The flowchart shown here explains the procedure of the characteristic image generation process in the fourth application example of the image processing device 10 .

[0337] In the following description, S701, S702, ... represent identification symbols of a plurality of steps in the feature image generation process of this application example. The feature image generation process of this application example starts from step S701.

[0338] <Steps S701 to S704>

[0339] The characteristic image generation unit 8c performs the same processing as steps S601 to S604 in steps S701 to S704. In step S704, if the characteristic image generation unit 8c has performed the processing of steps S701 to S703 on all of the plurality of size candidates, the processing proceeds to step S705.

[0340] <Steps S705, S706>

[0341] Furthermore, the characteristic image generating unit 8c executes the processes of steps S705 and S706, which are the same processes as steps S506 and S507. Thereafter, the characteristic image generating unit 8c ends the characteristic image generating process.

[0342] When this application example is adopted, the same effects as those of the first application example can be obtained.

[0343] [Fifth Application Example]

[0344] Next, the feature image generation process according to the fifth application example of the image processing device 10 will be described.

[0345] In this application example, the characteristic image generating unit 8c compares each pixel value of the first pre-processed image g11 and the second pre-processed image g12 with a predetermined reference range to distinguish pixels constituting the unique portion Ps1 from pixels not constituting the unique portion Ps1.

[0346] That is, in this application example, the characteristic image generating unit 8 c identifies the unique portion Ps1 based on the magnitude of each pixel value of the first pre-processed image g11 and the second pre-processed image g12 in the unique portion extraction process.

[0347] Furthermore, the characteristic image generating unit 8 c extracts the vertical line Ps11 by removing the unique portion Ps1 common to the first pre-processed image g11 and the second pre-processed image g12 from the unique portion Ps1 of the first pre-processed image g11 .

[0348] Furthermore, the characteristic image generating unit 8 c extracts the horizontal line Ps12 by removing the unique portion Ps1 common to the first pre-processed image g11 and the second pre-processed image g12 from the unique portion Ps1 of the second pre-processed image g12 .

[0349] Furthermore, the characteristic image generating unit 8 c extracts the unique portion Ps1 common to the first pre-processed image g11 and the second pre-processed image g12 as a noise point Ps13 .

[0350] For example, the characteristic image generating unit 8 c generates the first characteristic image g21 by converting the first pixel value Xi identified in the first pre-processed image g11 excluding the vertical line Ps11 into an interpolation value based on surrounding pixel values.

[0351] Similarly, the characteristic image generating unit 8 c generates a second characteristic image g22 by converting the second pixel values Yi identified in the second pre-processed image g12 excluding the horizontal line Ps12 into interpolated values based on surrounding pixel values.

[0352] Similarly, the characteristic image generating unit 8 c generates a third characteristic image g23 by converting the first pixel values Xi determined in the first pre-processed image g11 excluding the noise point Ps13 into interpolated values based on surrounding pixel values.

[0353] Alternatively, the characteristic image generating unit 8 c may generate the third characteristic image g23 by converting the second pixel values Yi identified in the second pre-processed image g12 excluding the noise point Ps13 into interpolation values based on surrounding pixel values.

Claims

1. An image processing method, characterized in that: The image processing method is an image processing method in which a processor determines that an image of a test image obtained by image reading processing of an output sheet of an image forming apparatus is defective. The image processing method comprises: The processor performs a first pre-processing with the horizontal direction of the test image as a processing direction, thereby generating a first pre-processed image, the first pre-processing including a main filtering process for transforming pixel values of pixels of interest sequentially selected from the test image into transformed values obtained by emphasizing a difference between pixel values of a region of interest including the pixel of interest and pixel values of two adjacent regions on both sides of the region of interest set in the processing direction; The processor generates a second pre-processed image by performing a second pre-processing including the main filtering process with the longitudinal direction of the test image as the processing direction; and The processor performs a special part extraction process, which extracts, from the special parts composed of one or more meaningful pixels in the first pre-processed image and the second pre-processed image, a first special part that exists in the first pre-processed image and is not common to the first pre-processed image and the second pre-processed image, a second special part that exists in the second pre-processed image and is not common to the first pre-processed image and the second pre-processed image, and a third special part that is common to the first pre-processed image and the second pre-processed image, respectively as the image defects.

2. The image processing method according to claim 1, wherein: The first pre-processing includes: performing the main filtering process with the horizontal direction as the processing direction, thereby generating first main mapping data; performing edge emphasis filtering processing on the test image with the horizontal direction as the processing direction and the target region and one of the two adjacent regions as targets, thereby generating horizontal edge intensity map data; and Correcting each pixel value of the first main map data using each pixel value of the corresponding horizontal edge intensity map data, thereby generating the first pre-processed image; The second pre-processing includes: performing the main filtering process with the longitudinal direction as the processing direction, thereby generating second main map data; performing the edge emphasis filtering process on the test image with the vertical direction as the processing direction and the target region and one of the two adjacent regions as targets, thereby generating vertical edge intensity map data; and The second pre-processed image is generated by correcting each pixel value of the second main map data using each pixel value of the corresponding vertical edge intensity map data.

3. The image processing method according to claim 1, wherein: In the special part extraction process, the processor derives an index value of the difference between corresponding pixel values in the first pre-processed image and the second pre-processed image, extracts the first special part by transforming each pixel value of the first pre-processed image using a first transformation formula predetermined based on the index value, extracts the second special part by transforming each pixel value of the second pre-processed image using a second transformation formula predetermined based on the index value, and extracts the third special part by transforming each pixel value of the first pre-processed image or the second pre-processed image using a third transformation formula predetermined based on the index value.

4. The image processing method according to claim 1, wherein: In the special part extraction process, the processor determines the special part based on the size of each pixel value of the first pre-processed image and the second pre-processed image, extracts the first special part by removing the special part common to the first pre-processed image and the second pre-processed image from the special part of the first pre-processed image, extracts the second special part by removing the special part common to the first pre-processed image and the second pre-processed image from the special part of the second pre-processed image, and extracts the special part common to the first pre-processed image and the second pre-processed image as the third special part.

5. The image processing method according to claim 1, wherein: The image processing method further includes the processor executing a compression process of compressing a read image obtained by the image reading process on the output sheet, thereby generating the test image.

6. The image processing method according to claim 5, characterized in that The processor compresses the read image using a plurality of compression ratios in the compression process, thereby generating a plurality of the test images having different sizes. Furthermore, the processor performs the first pre-processing and the second pre-processing on the plurality of test images, thereby generating a plurality of first pre-processed images and a plurality of second pre-processed images corresponding to the plurality of test images respectively. Furthermore, the processor extracts the first unique portion, the second unique portion, and the third unique portion through the unique portion extraction process based on the plurality of the first pre-processed images and the plurality of the second pre-processed images.

7. The image processing method according to claim 1, wherein: The processor performs a plurality of the first pre-processing and a plurality of the second pre-processing on one test image with different sizes of the target region and the adjacent region, thereby generating a plurality of the first pre-processing images and a plurality of the second pre-processing images. Furthermore, the processor extracts the first unique portion, the second unique portion, and the third unique portion through the unique portion extraction process based on the plurality of the first pre-processed images and the plurality of the second pre-processed images.

8. The image processing method according to claim 6, wherein: The processor performs the specific part extraction processing on multiple first pre-processed images and multiple second pre-processed images respectively, thereby extracting multiple candidates of the first specific part, the second specific part and the third specific part corresponding to the multiple test images, and then extracts the first specific part, the second specific part and the third specific part by aggregating the multiple candidates.

9. The image processing method according to claim 6, wherein: The processor aggregates multiple first pre-processed images and multiple second pre-processed images into one, performs the specific part extraction process on the aggregated first pre-processed images and second pre-processed images, thereby extracting the first specific part, the second specific part and the third specific part.

10. The image processing method according to claim 1, wherein: In the special part extraction process, the processor generates a first feature image obtained by extracting the first special part from the first pre-processed image, a second feature image obtained by extracting the second special part from the second pre-processed image, and a third feature image obtained by extracting the third special part from the first pre-processed image or the second pre-processed image.

11. The image processing method according to claim 10, wherein: The image processing method further includes: the processor executing a predetermined cause determination process using the first feature image, the second feature image, and the third feature image, thereby determining the causes of the first unique portion, the second unique portion, and the third unique portion.

12. The image processing method according to claim 11, wherein: The cause determination process includes periodic specific part determination process, The periodic specific part judgment processing is for the second characteristic image or the third characteristic image, to judge whether there is one or more predetermined periodicities in the longitudinal direction, and based on the judgment result of the periodicity, to judge the cause of the second specific part or the third specific part.

13. The image processing method according to claim 12, wherein: The cause determination process includes: a process of generating a non-periodic characteristic image by removing the second unique portion or the third unique portion synchronized with the periodicity from the second characteristic image or the third characteristic image; and The feature pattern recognition processing takes the non-periodic feature image as the input image, and through pattern recognition of the input image, determines which of the multiple predetermined cause candidates corresponding to the second specific part or the third specific part the input image corresponds to.

14. The image processing method according to claim 13, wherein: The feature pattern recognition process includes the following processes: The first feature image is used as the input image, and through the pattern recognition of the input image, it is determined to which of a plurality of predetermined cause candidates corresponding to the first unique portion the input image corresponds.

15. The image processing method according to claim 13, wherein: The feature pattern recognition process is a process of classifying the input image into any one of the plurality of cause candidates using a learning model obtained by previously learning a plurality of sample images corresponding to the plurality of cause candidates as training data.

16. The image processing method according to claim 11, wherein: The image processing method further includes: the processor determining a color vector, the color vector representing a vector in a color space from a color of the unique portion of the test image and a color of a reference area surrounding the unique portion toward another direction; The processor further uses the color vector in the cause determination process to determine the cause of the first unique portion, the second unique portion, or the third unique portion.

17. The image processing method according to claim 11, wherein: The image processing method also includes: the processor performs periodic unevenness judgment processing, and the periodic unevenness judgment processing judges the presence or absence of one or more predetermined periodicities in the longitudinal direction according to each predetermined color of the test image, and judges the presence and cause of periodic concentration unevenness as a type of image defect based on the periodicity judgment result.

18. The image processing method according to claim 17, wherein: The image processing method also includes: the processor determines whether the deviation of the pixel value exceeds a predetermined allowable range for each predetermined color for the test image determined to have no periodicity through the periodic unevenness judgment processing, thereby determining whether random concentration unevenness, which is a type of image defect, has occurred.

19. The image processing method according to claim 18, wherein: The image processing method also includes: the processor performs random pattern recognition processing, and the random pattern recognition processing uses the test image judged to have the random concentration unevenness as the input image, and through pattern recognition of the input image, determines which of one or more predetermined candidate causes of the image defect the input image corresponds to.

20. An image processing device, characterized in that: The image processing device includes a processor that executes the image processing method according to claim 1 .

Images