JP2008224745A - Image processing device, image processing program, malfunction diagnosing device, and malfunction diagnosis program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing device capable of improving accuracy in categorizing image defects, and to provide an image processing program, malfunction diagnosing device, and malfunction diagnosing program. <P>SOLUTION: A primary test chart is output (step 100), and an image defect area of the read primary test chart image is detected (step 104). The color component of an arising image defect and the type of the defect are categorized (step 106). If the color component and the type cannot be specified (No, in step 110), the degree of similarity of the defect that cannot be specified is calculated (step 114). Based on the degree of similarity, parameters are altered (step 106). Then, a secondary test chart determined based on the altered parameter is output (step 120), and the color component of an image defect arising in the secondary test chart and the type of the defect are categorized (step 126). <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an image processing device, an image processing program, a failure diagnosis device, and a failure diagnosis program.

  2. Description of the Related Art Conventionally, periodic maintenance by a professional service person, diagnosis and repair of a fault location, and the like have been performed on image forming apparatuses such as copiers and printers. However, with the recent trend toward colorization and higher functionality of image forming apparatuses, failure modes have become more complex, and even a professional service person may not be able to identify the cause of the failure. In such a case, since it is necessary to minimize the downtime of the image forming apparatus, a plurality of parts that are likely to be related to the cause of the failure are replaced together. For this reason, there is a problem that normally normal parts are also exchanged together, resulting in an increase in the cost required for failure repair.

  In order to solve this problem, a technique for printing a test chart (inspection target image) and diagnosing a failure portion of the image forming apparatus from an image defect state of the test chart is known. In this technique, when an image defect occurs, a test chart corresponding to the state of the generated image defect is selected from a plurality of types of test charts stored in the image forming apparatus, and image defect diagnosis is performed. . In the image defect diagnosis, the selected test chart is output, the image defect generated in the output test chart is detected, the type of the image defect is classified, and the image forming apparatus determines the type of the image defect based on the classification result and the generation level of the image defect. Estimate the failure location.

  As such technology, the machine data and job data of the image forming device are collected, the collected data is analyzed by the diagnostic reasoning engine, the initial diagnosis result is obtained, and the test chart is determined and output based on the initial diagnosis result The technique to do is known (for example, refer patent document 1).

Also, a dynamic test pattern generation technique that automatically selects an optimal test pattern set from a plurality of test patterns (test charts) is known (see, for example, Patent Document 2). In this technique, an initial diagnosis is performed from internal information of the machine and operator input information, and a test pattern is selected from the initial diagnosis result and output. Then, image quality analysis is performed on the output test pattern, the feature of the defect is extracted, the certainty of the diagnosis result by the diagnosis engine is calculated, and if the certainty level is less than the threshold value, the previously associated defect The test pattern set corresponding to the defect is automatically selected from the relationship between the feature and the test pattern, and the image quality analysis and the diagnosis process are executed again.
JP 2001-245091 A JP 2006-14332 A

  By the way, when only a test chart stored in advance in the image forming apparatus is used, among image defects such as density unevenness and density too low (LoID), those with a small degree are classified in the image defect diagnosis. Is difficult.

  Among specific defects generated in an electrophotographic image forming apparatus, a specific example in which it is difficult to classify image defects is a test chart (FIG. 27A) shown in FIG. A classification of the generated CMY color loss and K single color loss is given. In order to calculate the feature of the defect for the broken line image defect area in FIG. 27, the projection waveform in the main scanning direction is calculated and classified, but in the case of CMY color loss (FIG. 27B) and K single color. In the case of missing (FIG. 27C), the characteristics of the RGB components (FIG. 27D) of the scanned image are the same, and it is difficult to distinguish which image defect has occurred.

  As another example, FIG. 28 shows the classification of density unevenness called InOut density unevenness and LoID with a small degree of image defect. In the case of InOut density unevenness and LoID having a small degree of image defect shown in FIG. 28, the occurrence of defects is similar, and therefore, as shown in FIG. 29, InOut density unevenness is plotted even if plotted in a feature space (details will be described later). It will be plotted at a position that cannot be determined from either the space (details will be described later) or the LoID space (details will be described later), and it is difficult to specify the type of image defect.

  Furthermore, FIG. 30 shows a classification of density unevenness and lines / streaks called banding (streaky density unevenness) with a small degree of image defect. In the case of banding, it is important to extract the periodicity of banding. However, if the degree of banding is small as shown in FIG. 30, the periodicity cannot be extracted, and the characteristics are almost the same as in the case of the horizontal line. . Therefore, even if plotted in the feature space, it is plotted at a position that cannot be determined from either the banding space (detailed later) or the horizontal line space (detailed later), and it is difficult to specify the type of image defect.

  An object of the present invention is to provide an image processing device, an image processing program, a failure diagnosis device, and a failure diagnosis program that can improve the classification accuracy of image defects.

  In order to achieve the above object, an image processing apparatus according to claim 1 is configured to output a primary inspection target image output unit that outputs a predetermined primary inspection target image, and the primary inspection target image output unit that outputs the primary inspection target image. First defect area detecting means for detecting an image defect area in which an image defect has occurred from image data of an image obtained by reading an image to be subjected to a primary inspection, and a first classifying image defect occurring in the image defect area Similarity calculation for calculating the similarity between each of the generated image defects and a plurality of predetermined image defects according to the classification result of the first image defect classification means and the first image defect classification means A parameter changing unit that changes a parameter relating to a primary inspection target image based on the classification result of the first image defect classification unit and the similarity calculated by the similarity calculation unit; A secondary inspection target image output means for outputting a secondary inspection target image based on the parameter changed by the data changing means, and an image obtained by reading the secondary inspection target image output from the secondary inspection target image output means Second defect area detecting means for detecting an image defect area where an image defect has occurred from the image data of the second image, and a second image for classifying the image defect occurring in the image defect area of the secondary inspection target image Defect classification means.

  The image processing device according to claim 2 is the image processing device according to claim 1, wherein the first image defect classification means includes color components of image defects and image defects generated in the primary inspection target image. The second image defect classification means classifies the color component of the image defect occurring in the secondary inspection target image and the type of the image defect.

  The image processing device according to claim 3 is the image processing device according to claim 1 or 2, wherein the similarity calculation unit calculates a feature value of an image defect occurring in the primary inspection target image. Then, the similarity is calculated based on the relationship between the calculated feature value and the feature value predetermined for each of the plurality of types of image defects.

  An image processing apparatus according to a fourth aspect is the image processing apparatus according to the third aspect, wherein the feature value includes a size, a density, and a frequency component of the image defect.

  An image processing apparatus according to a fifth aspect is the image processing apparatus according to the third or fourth aspect, wherein the similarity calculation unit calculates a difference between characteristic values determined in advance for the plurality of types of image defects. Is to be calculated.

  The image processing apparatus according to claim 6 is the image processing apparatus according to any one of claims 1 to 5, wherein the parameter changing unit includes at least a toner density and a color component of the primary inspection target image. One is to change.

  The failure diagnosis apparatus according to claim 7, wherein a feature amount extraction unit that extracts a feature amount related to a degree of image defects occurring in the secondary inspection target image, and a feature amount extracted by the feature amount extraction unit, The cause of the occurrence of the image defect is specified as the cause of failure based on the classification result of the second image defect classification means of the image processing apparatus according to any one of claims 1 to 6. Specifying means.

  The fault diagnosis apparatus according to claim 8 is the fault diagnosis apparatus according to claim 7, wherein the specifying unit selects the classification result from a plurality of Bayesian networks provided for each kind of predetermined image defect. The Bayesian network selected according to the method is used.

  The failure diagnosis apparatus according to claim 9 is the failure diagnosis apparatus according to claim 7 or 8, further comprising display means for displaying a specific result of the specifying means.

  The image processing program according to claim 10, wherein a predetermined primary inspection target image is output, and an image defect in which an image defect is generated from image data of an image obtained by reading the output primary inspection target image A step of detecting an area, a step of classifying an image defect occurring in the image defect area, an image defect occurring according to the classification result of the image defect, and a plurality of predetermined types of image defects Calculating the degree of similarity with each other, changing the parameter relating to the primary inspection target image based on the classification result of the image defect, and the degree of similarity, and the secondary inspection target based on the changed parameter Outputting an image, and detecting an image defect area in which an image defect has occurred from the image data of the read image of the secondary inspection target image output Step a, to perform the steps of classifying an image defect occurring in the image defect area of the secondary target image, the processing including a computer.

  The failure diagnosis program according to claim 11 is a step of extracting a feature amount related to a degree of an image defect occurring in the secondary inspection target image, the extracted feature amount, and the claims 1 to. And a step of identifying a cause of the occurrence of the image defect as a cause of failure based on the classification result of the second image defect classification means of the image processing apparatus according to any one of claims 6 to 6. To run.

  As described above, according to the present invention described in claims 1 to 6, it is possible to improve the image defect classification accuracy.

  According to the seventh and eighth aspects of the present invention, it is possible to improve the accuracy of the failure diagnosis that identifies the cause of the failure that causes the image defect.

  According to the present invention as set forth in claim 9, it is possible to notify the user of the failure diagnosis result.

  According to the tenth aspect of the present invention, there is an effect that it is possible to cause a computer to execute processing for classifying image defects with high accuracy.

  According to the eleventh aspect of the present invention, there is an effect that it is possible to cause the computer to execute the process of performing the fault diagnosis with high accuracy.

[First Embodiment]
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 shows a schematic configuration of an image forming apparatus 10 including an image processing apparatus and a failure diagnosis apparatus according to the present embodiment. The image forming apparatus 10 according to the present embodiment includes a network I / F (network interface) unit 12, a print engine unit 14, an image reading unit 16, an operation instruction unit 18, a user interface unit 20, a storage unit 22, and a test chart storage unit. 24, a CPU 26, an ASIC (application specific integrated circuit) 28, a loop counter 30, an image processing unit 32, and a failure diagnosis unit 34.

  The network I / F unit 12 is a network interface for communicating print data (image data) 11 instructed to be printed by a user or the like. The print engine unit 14 forms an image instructed to be printed on the recording medium 15 and outputs it. In the present embodiment, specifically, it is an electrophotographic printer, which uses C (cyan) color, M (magenta) color, Y (yellow) color, or K (black) color toner or ink. An image is formed on the recording medium 15. In addition, when forming an achromatic color on the recording medium 15, there are a case where it is formed using only the K color and a case where it is formed by mixing the C color, the M color, and the Y color.

  The image reading unit 16 reads an image (test chart) formed on the recording medium 15 and outputs image data, and is, for example, a scanner.

  For example, the operation instruction unit 18 issues an operation instruction related to image formation by a user operation. The user interface unit 20 notifies the user of information relating to image formation, and is, for example, a liquid crystal display.

  The storage unit 22 includes a memory, a hard disk, a ROM, a RAM, and the like (not shown), a program for executing each processing flow, a default value of the loop counter 30 (the number of times of loop processing), a parameter change rule , The relationship between the toner density and the similarity R, the center value vector of each type of image defect in the feature space, the threshold for specifying the color component, the threshold for specifying the defect type (similarity), and the Bayesian network (respectively, Etc.) are stored in advance. The test chart storage unit 24 is, for example, a nonvolatile storage device such as a hard disk, and stores various test charts (inspection target images) in advance.

  The CPU 26 and the ASIC 28 control the entire image forming apparatus 10. The loop counter 30 is for counting the number of executions of loop processing, which will be described in detail later.

  The image processing unit 32 corresponds to the image processing apparatus of the present embodiment, and is generated in the test chart based on the image data of the test chart output from the print engine unit 14 and read by the image reading unit 16. Classification of image defects. The failure diagnosis unit 34 corresponds to the failure diagnosis apparatus of the present embodiment, and diagnoses a failure of the image forming apparatus 10 based on the image defects classified by the image processing apparatus.

  Network I / F unit 12, print engine unit 14, image reading unit 16, operation instruction unit 18, user interface unit 20, storage unit 22, test chart storage unit 24, CPU 26, ASIC 28, loop counter 30, image processing unit 32, In addition, the failure diagnosis unit 34 is connected to be able to exchange information with each other via a bus 36 such as a control bus or a data bus.

  Next, the image processing apparatus 50 according to the present embodiment will be described with reference to FIG. FIG. 2 is a functional block diagram illustrating a schematic configuration of the image processing apparatus 50.

  The image processing apparatus 50 according to the present embodiment includes an image processing unit 32, a test chart storage unit 24, a print engine unit 14, an image reading unit 16, a loop counter 30, and a storage unit 22.

  The image processing unit 32 includes a primary test chart output unit 52, a first image defect area detection unit 54, a first image defect classification unit 56, a first image defect classification end determination unit 57, a similarity calculation unit 58, and a parameter change. Means 60, secondary test chart determination means 62, secondary test chart output means 64, second image defect area detection means 66, second image defect classification means 68, and second image defect classification end determination means 69; Prepare.

  The primary test chart output unit 52 instructs the print engine unit 14 to output the test chart read from the test chart storage unit 24. The primary test chart output means 52 is connected to the first image defect area detection means 54, and the first image defect area detection means 54 outputs the image data of the primary test chart output from the image reading unit 16. An image defect area is detected. The first image defect classification means 56 is connected to the first image defect area detection means 54, and the first image defect classification means 56 generates an image defect that has occurred with respect to the detected image defect area. Classify. In the present embodiment, the color components and types of image defects are classified. The first image defect classification means 56 is connected to first image defect classification end determination means 57. The first image defect classification end determination means 57 is a means for determining whether the value of the loop counter 30 or the first image defect classification means 56 has specified one type of image defect. The similarity calculation unit 58 calculates the similarity based on the classification result of the image defect classification process. A parameter changing means 60 is connected to the similarity calculating means 58, and the parameter changing means 60 changes the parameters of the primary test chart. A secondary test chart determining unit 62 is connected to the parameter changing unit 60, and the secondary test chart determining unit 62 determines a secondary test chart based on the changed parameter. A secondary test chart output unit 64 is connected to the secondary test chart determination unit 62, and the secondary test chart output unit 64 outputs the determined secondary test chart to the print engine unit 14. Instruct. The secondary image chart output means 64 is connected to the second image defect area detection means 66, and the second image defect area detection means 66 is an image of the secondary test chart output from the image reading unit 16. Detect image defect areas in the data. The second image defect classification means 68 is connected to the second image defect area detection means 66, and the second image defect classification means 68 is connected to the detected image defect area. Classify. The second image defect classification end judging means 69 is a means for judging whether the value of the loop counter 30 or the second image defect classification means 68 has specified one type of image defect. In the present embodiment, the color components and types of image defects are classified. In addition, a failure diagnosis device 70 is connected to the second image defect classification end determination means 69, and the identified image defect classification result is output to the failure diagnosis device 70. Note that the user interface unit 20 may notify the user of the image defect classification result.

  Furthermore, the failure diagnosis apparatus 70 according to the present embodiment will be described with reference to FIG. FIG. 3 is a functional block diagram illustrating a schematic configuration of the failure diagnosis apparatus 70. The failure diagnosis apparatus 70 according to the present embodiment includes a failure diagnosis unit 34.

  The failure diagnosis unit 34 includes a feature amount extraction unit 72, a feature amount calculation unit 74, a failure identification unit 76, and a storage unit 22. The feature amount extraction unit 72 extracts the feature amount of the generated image defect as information. For example, for an image defect in which a line is generated, information on the width and periodicity of the generated line is provided. Is a line extraction filter or the like. A feature quantity calculation unit 74 is connected to the feature quantity extraction unit 72, and the feature quantity value calculation unit 74 calculates the feature quantity of the image defect based on the image defect information extracted by the feature quantity extraction unit 72. To do. A failure specifying unit 76 is connected to the feature amount calculating unit 74, and the failure specifying unit 76 includes the feature amount calculating unit 74 based on the calculated feature amount and the classification of the image defect output from the image processing apparatus 50. Identify the cause of failure. The result specified by the failure specifying means 76 (failure diagnosis result) may be notified to the user via the user interface unit 20.

Next, image processing executed by the image processing apparatus 50 according to the present embodiment and failure diagnosis processing executed by the failure diagnosis apparatus 70 will be described in detail with reference to FIGS.
(Image defect diagnosis processing)
First, the image defect diagnosis process will be described. The image defect diagnosis process of the present embodiment is a process including a process of classifying image defects executed by the image processing device 50 and a failure diagnosis process executed by the failure diagnosis device 70. FIG. 4 shows a flowchart of the image defect diagnosis process of the present embodiment. The image defect diagnosis process shown in FIG. 4 is executed when the user gives an instruction via the operation instruction unit 18 when an image defect occurs in the image of the recording medium output from the print engine unit 14, for example. The

  First, in step 100, the print engine unit 14 is instructed to read the primary test chart from the test chart storage unit 24 and output the read primary test chart. Note that a general test chart can be used as the primary test chart of the present embodiment. For example, a test chart in which CMYK is printed in solid.

  In the next step 102, the primary reading chart output from the print engine unit 14 is instructed to be read by the image reading unit 16. In the next step 104, an image defect area of the read image data of the primary test chart is detected. The detection of the image defect area according to the present embodiment is performed by, for example, performing binarization processing on the read image data of the primary test chart, and setting the area where the gradation value of the binary image is 0 or 1 as the image defect area Detect as. Further, by performing label processing on all detected image defect areas, image defects in units of image defect areas may be detected.

  In the next step 106, an image defect classification process is executed. The image defect classification process is a process of determining whether the color component in which an image defect has occurred and the type of image defect can be specified as one type and classifying them (details will be described later).

  When the image defect classification process is completed, in the next step 108, it is determined whether or not the loop counter 30 is equal to or greater than a predetermined value, that is, whether or not a predetermined number of loop processes have been executed. If (the loop processing has been executed a predetermined number of times), an affirmative determination is made and the routine proceeds to step 130. On the other hand, if the loop counter 30 is less than the predetermined value (the loop processing has not yet been executed a prescribed number of times), the process proceeds to step 110. In the present embodiment, the predetermined value is stored in the storage unit 22 in advance.

  In step 110, it is determined whether or not one type of image defect occurring in the test chart is specified. If one type of image defect is specified by the image defect classification process, an affirmative determination is made, and the routine proceeds to step 130. On the other hand, if it is not specified as one type, a negative determination is made and the routine proceeds to step 112.

  In step 112, it is determined whether or not the color component of the image defect occurring in the test chart is specified. If it is determined by the image defect classification process that the color component of the image defect cannot be specified, a negative determination is made and the process proceeds to step 116. On the other hand, if the color component is specified, an affirmative determination is made and the routine proceeds to step 114.

  In step 114, similarity calculation processing is performed. The similarity calculation process is to calculate the similarity R of image defects of a plurality of types (two types in the present embodiment) that cannot be specified when the type of image defects occurring in the test chart cannot be specified as one type. (Details will be described later).

  In the next step 116 when the similarity calculation processing is completed and when a negative determination is made in step 112, the type of image defect that cannot be specified, the similarity R of the image defect, and the color component cannot be specified (color component unknown). The parameters of the primary test chart are changed according to the determination result. In the present embodiment, the parameter is changed based on a parameter change rule that is set in advance and stored in the storage unit 22. An example of the parameter change rule is shown in FIG.

  As shown in FIG. 5, when the color component in which the defect area has occurred is unknown, two types of parameters, parameter 1 and parameter 2, are defined and changed to any one of the parameters. Parameter 1 is to change the color components of the primary test chart so as to output only YMC colors, and parameter 2 is to output the color components of the primary test chart only with K colors. In addition, when it is impossible to specify which defect of density too low (LoID) or InOut density unevenness has occurred, parameter 1 shows the toner density (all colors) of the primary test chart based on the similarity R. In parameter 2, the toner density of the color component specified in the primary test chart is changed according to the relationship shown in FIG. 6 and 7 are diagrams illustrating an example of the relationship (parameter curve) between the similarity R and the toner density change amount, which are stored in the storage unit 22 in advance. In any case, the amount of change in toner density (the slope of the parameter curve) varies depending on the type of image defect. The smaller the similarity R, that is, the more similar the image defect that cannot be identified, the higher the toner density of the primary test chart. In this way, the toner density of the primary test chart is changed to be higher as the degree of similarity is lower and the degree of image defect cannot be specified increases.

  Further, when it is impossible to determine which defect of line / streaks or banding density unevenness has occurred, the toner density (all colors) of the primary test chart is changed according to the relationship shown in FIG. (Parameter 1)

  In the next step 118, a secondary test chart is determined. In the present embodiment, the primary test chart output in step 100 is read from the test chart storage unit 24 and the parameter change is determined as a secondary test chart. In the next step 120, the determined secondary test chart is determined. The print engine unit 14 is instructed to output the chart. In the next step 122, the image reading unit 16 is instructed to read the output secondary test chart, and the process proceeds to step 124.

  In step 124, an image defect area of the read image data of the secondary test chart is detected. Note that this image defect area detection may be performed in the same manner as the image defect area detection in step 104, and therefore detailed description thereof is omitted here.

  In the next step 126, an image defect classification process is performed. Since this image defect classification process is the same as the image defect classification process in step 106, detailed description thereof is omitted here.

  In the next step 128, the loop counter 30 is incremented, and the process returns to the determination of whether or not the loop counter 30 in step 108 is equal to or greater than a predetermined value, that is, whether or not the specified number of loop processes have been executed. As a result, the loop counter 30 becomes equal to or larger than the predetermined value (a predetermined number of loop processes are executed), or the color component of the image defect occurring in the test chart is specified and the type of the image defect is specified as one type. Step 108 to Step 128 are repeated until it is done.

  Then, in the next step 130 in the case where the loop counter 30 is equal to or larger than the predetermined value (Yes in Step 108) or the type of image defect is specified as one type (Yes in Step 110), failure diagnosis processing is performed. The failure diagnosis process is a process for diagnosing a failure of the image forming apparatus 10 based on the image defect classification result (details will be described later).

After the failure diagnosis process in step 130 is finished, the present process (image defect diagnosis process) is finished.
(Image defect classification processing)
Next, the image defect classification process (the process of step 106 of the image formation diagnosis process shown in FIG. 4) will be described in detail with reference to FIGS. FIG. 8 is a flowchart illustrating an example of the image defect classification process.

  First, in order to specify which color component of the R component, G component, and B component has an image defect, in step 200, the image data of the primary test chart is decomposed into RGB color components. Then, in the next step 202, in the feature space, the reciprocal numbers Dist_R, Dist_G, and Dist_B of the feature values of the RGB color components and the distance between the origins are calculated by Equations (1) to (3).

In the present embodiment, the feature space is a three-dimensional space using the image defect size S, the frequency component F, and the density D as parameters. FIG. 9 shows an example of the distance between the feature value of each RGB color component and the feature value of the R component and the origin in the feature space.

  Further, in the next step 204, the smallest value is extracted from Dist_R, Dist_G, and Dist_B as Dist_defect1, the next smallest one is extracted as Dist_defect2, and the next smallest (maximum) is extracted as Dist_defect3.

  In the next step 206, image defect color component identification determination processing is performed. The image defect color component identification determination process is a process for determining whether one color component in which an image defect has occurred can be identified (details will be described later).

  When the image defect color component identification determination process is completed, in the next step 208, it is determined whether or not the color component in which the image defect has occurred is identified. If the result of the image defect color component identification determination process in step 206 is determined that one type of color component cannot be identified, a negative determination is made and the process proceeds to step 210. In step 210, an image defect has occurred. It is determined that it is impossible to specify a color component (color component is unknown), and this processing is terminated.

  On the other hand, if the result of the image defect color component identification determination process determines that the color component in which the image defect has occurred can be identified, it is determined in step 208 that the color component is identified (affirmative determination). Then, the process proceeds to step 212.

  In order to specify the type of the image defect that has occurred for the specified color component by clustering processing, in step 212, the feature of the defect area of the color component in which the image defect has occurred (the specified color component) As the center value vector of the values, the image area size Sdef of the defect area, the average density Ddefection of the defect area, and the peak value Fdef of the frequency component are calculated.

  In the next step 214, the distance between the feature value of the image defect area and the feature value of each type of image defect is calculated. In the present embodiment, an image defect space corresponding to each type of image defect is predetermined in the feature space. For example, there are InOut density unevenness space, LoID space, banding space, horizontal line space, and the like. The center value vector of these image defect spaces is stored in the storage unit 22 in advance. For example, when the type of defect is “point (a point image that does not originally occur in the test chart)”, the center value vector of the feature value is represented as (Sspot, Dspot, Fspot). The Euclidean distance Dist_Spot between the center value vector of the “point” defect and the center value vector (Sdef, Ddef, Fdef) of the defect area of the test chart is calculated by the equation (4).

Further, the Euclidean distance between the center value vector of another type of defect and the center value vector of the defect area of the test chart is calculated in the same manner. For example, when the defect type is InOut density unevenness, the Euclidean distance Dist_InOut is calculated from the feature value center value vectors (Sinout, Dinout, Finout) stored in the storage unit 22 in advance. Further, for example, when the defect type is too low in density (LoID), the Euclidean distance Dist_LoID is calculated from the feature value center value vectors (Sloyd, Fluid, Fluid) stored in advance in the storage unit 22.

  Although the Euclidean distance is calculated in the present embodiment, the Mahalanobis distance may be calculated. In this case, a central value vector of feature values of each type of image defect necessary for Mahalanobis distance calculation is stored in the storage unit 22 in advance.

  In the next step 216, the smallest one of Dist_Spot, Dist_LoID, and Dist_InOut, that is, the one having the shortest distance from the center value vector of the defect area of the test chart is set as Dist_defect1 ′, and the next smallest one (short distance) ) Is extracted as Dist_defect2 ′, and the largest one (long-distance) is extracted as Dist_defect3 ′. In this embodiment, as an example, since classification is made into any of three types of image defects, Dist_defects 1 ′ to 3 ′ are extracted here, but any one of more than three types of image defects is selected. However, the classification is not limited to this.

  In the next step 218, an image defect type identification determination process is performed. The image defect type specific determination process is a process of determining whether or not one type of generated image defect can be specified (details will be described later).

  When the image defect color component identification determination process ends, in the next step 220, it is determined whether or not the type of image defect can be identified as one type. If the result of the image defect type identification determination process in step 216 determines that the type of image defect cannot be identified as one type, a negative determination is made, and the process proceeds to step 222. In step 222, the generated image is detected. It is determined that it is impossible to specify the type of defect, and this process is terminated.

On the other hand, when the result of the image defect type identification determination process determines that one type of image defect can be identified, it is determined in step 220 that the type of image defect is identified as one type (Yes) Determination), and the process proceeds to step 224. In step 224, it is determined that the type of the image defect that has occurred can be specified, and the present process ends.
(Image defect color component identification determination process)
Next, the image defect color component identification determination process (the process of step 206 of the image defect classification process shown in FIG. 8) will be described in detail with reference to FIG. FIG. 10 is a flowchart illustrating an example of the image defect color component identification determination process.

  First, in step 300, equation (5) is calculated.

D1 = | Dist_defect1-Dist_defect2 | (5)
Further, in the next step 302, equation (6) is calculated.

D2 = | Dist_defect1-Dist_defect3 | (6)
In the next step 304, the threshold value stored in advance in the storage unit 22 is read, and it is determined whether or not at least one of D1 and D2 is equal to or greater than the threshold value. If both D1 and D2 are less than the threshold value, a negative determination is made, and the process proceeds to step 306. In step 306, after determining that it is impossible to specify one type of color component in which an image defect has occurred, this processing is performed. Exit.

On the other hand, if at least one of D1 and D2 is greater than or equal to the threshold value, an affirmative determination is made, and the routine proceeds to step 308. In step 308, after determining that the image defect color component can be specified as one type, the present process is terminated. In the present embodiment, when only D1) is greater than or equal to the threshold, the color component corresponding to Dist_defect2 is selected. When only D2 is greater than or equal to the threshold, the color component corresponding to Dist_defect3 is greater than D1 and D2. In this case, the color components corresponding to Dist_defect1 are specified as image defect color components, respectively.
(Image defect type identification process)
Next, the image defect type identification determination process (the process of step 218 of the image defect classification process shown in FIG. 8) will be described in detail with reference to FIG. FIG. 11 is a flowchart illustrating an example of the image defect type identification determination process.

First, at step 400, equation (7) is calculated.
D3 = | Dist_defect1′−Dist_defect2 ′ | (7)
In the next step 402, the threshold value stored in advance in the storage unit 22 is read, and it is determined whether D3 is equal to or greater than the threshold value. If it is less than the threshold value, a negative determination is made, and the process proceeds to step 404. In step 404, it is determined that the type of the image defect that has occurred cannot be specified as one type, and then this process ends.

On the other hand, if D3 is greater than or equal to the threshold value, an affirmative determination is made, and the routine proceeds to step 406. In step 406, after it is determined that the type of the image defect that has occurred can be specified as one type, this processing is terminated. In the present embodiment, when D3 is equal to or larger than the threshold, the type of defect corresponding to Dist_defect1 ′ is specified as the type of image defect occurring in the test chart.
(Similarity calculation processing)
Next, the similarity calculation process (the process of step 114 of the image defect diagnosis process shown in FIG. 4) will be described in detail with reference to FIGS. FIG. 12 shows an example in which it is difficult to classify the image defect defect1 ′ and the image defect defect2 ′. Note that the image defect defect1 ′ has a center value vector (Sdefect1 ′, Ddefect1 ′, Fdefect1 ′) that is closest to the center value vector of the defect area of the test chart in the image defect classification process (distance Dist_defect1 ′). The type of image defect (for example, InOut density unevenness), and the type of image defect having the center value vector (Sdefect2 ′, Ddefect2 ′, Fdefect2 ′) that is next closest to the image defect defect2 ′ (distance Dist_defect2 ′) (For example, LoID).

  FIG. 13 is a flowchart illustrating an example of similarity calculation processing. In step 500, the absolute value of the difference in distance between the center value vector of the defect feature value of the test chart and the center value vector of each image defect is calculated as the similarity according to the equation (8), and this processing is terminated.

Similarity R = | Dist_defect1′−Dist_defect2 ′ | (8)
Note that the closer the value between the distance Dist_defect 1 ′ and the distance Dist_defect 2 ′ is, the closer the similarity R is to 0, and conversely, the greater the value is, the greater the similarity R is. Accordingly, as the similarity R is closer to 0 (similarity R is smaller), the image defects def1 ′ and image defect def2 ′ are more similar to each other, and the type of the image defect occurring is the image defect def1 ′ and This means that it is difficult to classify the image defect defect2 ′.
(Failure diagnosis processing)
Next, the failure diagnosis process (the process of step 130 of the image defect diagnosis process shown in FIG. 4) will be described with reference to FIGS. In the present embodiment, as an example of failure diagnosis processing, failure diagnosis processing using a Bayesian network is performed. FIG. 14 is a flowchart showing failure diagnosis processing. In step 600, the feature amount of the specified image defect is calculated. For example, an image defect in which a black line occurs in the test chart has occurred, and the image processing device 50 classifies the color component in which the image defect has occurred into K (black) color and classifies the defect type as a line. If there is, the feature amount extraction unit 72 such as a line or point extraction filter calculates the line width or periodicity of the black line as the feature amount based on the extracted information.

  In the next step 602, a Bayesian network corresponding to the type of image defect is read from the storage unit 22. In the present embodiment, a Bayesian network corresponding to a plurality of types of image defects is stored in the storage unit 22 in advance. A Bayesian network is a problem area where causal relationships are complex. Therefore, causal relationships between multiple variables associated with conditional probability distributions are sequentially connected and expressed as a network having a graph structure. The dependency relationship between them is represented by a directed graph. That is, a Bayesian network is a model of problem areas using establishment theory (for details, refer to Japanese Patent Laid-Open No. 2005-309078). As an example of the stored Bayesian network, FIG. 15 shows a Bayesian network corresponding to an image result in which a black line is generated. As shown in FIG. 15, the nodes are connected so as to have a relationship of “cause” → “result”. For example, the relationship between “drum scratches” and “line width information” is such that “line width information” such as “drum scratches” originated (cause) and thin lines (results) appeared. On the other hand, the relationship between “feed number history information” and “fuser” is based on the “number of feeds” (the number of feeds is more than one), and there is a possibility of black line generation due to “fuser” degradation. The relationship of becoming higher holds. The initial value of the probability data of each node is determined based on past data, for example. After that, the probability of each node may be periodically updated based on market trouble statistical data such as the frequency of replacement of parts of the image forming apparatus 10 and the frequency of occurrence of defects. Further, the state of the node representing the feature of the image defect such as “line width information” and “periodicity information” indicated by hatching in FIG. 15 is determined by the feature amount calculated in step 600.

  In the next step 604, the cause of the failure is identified based on the read Bayesian network, and this process is terminated. The specified cause of failure may be displayed to the user by the user interface unit 20.

  In the present embodiment, the case where the image processing apparatus 50 identifies the color component in which an image defect has occurred and the type of defect is classified into one type has been described. In the case of classification, failure diagnosis processing may be performed using a Bayesian network to which information according to a plurality of classified defects is added.

  In addition, the failure diagnosis processing is not limited to this. For example, a diagnosis processing based on a rule (for example, a table or the like) in which a failure portion corresponding to the image defect type and the feature amount of the defect area is described, or a neural network is used. It may be a diagnostic process or the like.

  As a result, failure diagnosis of the image forming apparatus 10 can be performed based on the image defects classified by the image processing apparatus 50. Further, since the failure diagnosis can be performed based on the result of the image defect classification with the image processing apparatus 50 with high accuracy, the failure diagnosis can be performed with high accuracy. For example, in the case of using a Bayesian network, if an image defect is classified incorrectly, it will lead to an incorrect cause of failure, but this can be prevented by correctly classifying it. Failure diagnosis can be performed with high accuracy.

  As described above, in the present embodiment, the similarity R of a plurality of image defects occurring in the primary test chart is calculated, the parameters of the primary test chart are changed based on the similarity R, and the secondary test chart is changed. Since the test chart is output, the image defects can be classified based on the secondary test chart. Thereby, since it is possible to classify the image defect using the secondary test chart whose parameters are changed according to the image defect that has occurred, it is possible to classify the image defect with high accuracy. Therefore, it is possible to classify which color components and types of image defects have occurred for image defects that are difficult (similar) to be classified.

  In addition, since image defects occurring in the test chart can be classified with high accuracy, failure diagnosis of the image forming apparatus 10 can be performed with high accuracy based on the classification result.

[Second Embodiment]
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the present embodiment, as a specific example, a case will be described in which InOut concentration unevenness actually occurs, but it is difficult to classify the concentration as too low (LoID). Since the present embodiment has substantially the same configuration and processing as the first embodiment, the same parts are denoted by the same reference numerals and detailed description thereof is omitted.

  First, when the image defect diagnosis process (see FIG. 4) is started, in step 100, C (cyan) 30%, M (magenta) 30%, Y (yellow) 30%, K (black) 30 as primary test charts. The print engine unit 14 is instructed to output a test chart printed with being superimposed on the entire surface of the image forming area on the recording medium with a toner density of%. The primary test chart after output is shown in FIG. In the present embodiment, it is assumed that an image defect has occurred on the right side of the primary test chart in the main scanning direction.

  In the next step 102, the image reading unit 16 is instructed to read the output primary test chart, and the process proceeds to step 104. In step 104, an image defect area is detected. FIG. 17 shows the detection result of the image defect area.

  In the next step 106, image defect classification processing is performed. FIG. 18 shows plot positions of feature values extracted from the center value vector in the preset InOut density unevenness on the feature space, the center value vector in the preset LoID, and the image defect area of the primary test chart. The plot position on the feature space in the primary test chart is located at the closest distance from the center value vector of the InOut density unevenness, but is located at the same distance from the LoID center value vector. And the LoID, the type of image defect cannot be specified as one type. Therefore, as a result of the image defect classification process, it is determined that one type of image defect cannot be specified.

  Therefore, if a negative determination is made in the next step 108, a negative determination is made in the next step 110, and if an affirmative determination is made in the next step 112, the process proceeds to step 114. In step 114, the image defect classification process is performed. Based on the result, similarity calculation processing is performed.

  Further, in the next step 116, the parameters of the test chart are changed. The parameter is changed based on a parameter change rule set in advance and stored in the storage unit 22. In the present embodiment, the table shown in FIG. 5 is used as the parameter change rule. In this embodiment, since it cannot be specified as either InOut density unevenness or LoID, the parameter to be changed is the toner density, and the change amount is based on the similarity R according to the relationship of FIG. It is determined.

  Based on the parameter changed in the next step 118, the secondary test chart is determined. In the next step 120, the print engine unit 14 is instructed to output the determined secondary test chart. An image of the output secondary test chart is shown in FIG. In the next step 122, the image reading unit 16 is instructed to read the output secondary test chart, and the process proceeds to step 124.

  In step 124, the output secondary test chart image defect area is detected, and in the next step 126, image defect classification processing is performed. FIG. 20 shows plot positions of feature values extracted from the center value vector in the preset InOut density unevenness on the feature space, the center value vector in the preset LoID, and the image defect area of the primary test chart. Since the plot position on the feature space in the secondary test chart is located closer to the center value vector of the InOut density unevenness than the center value vector of the LoID, as a result of the image defect classification process, the image defect type is changed to the InOut density unevenness. Identified.

  Thus, after the loop counter 30 is incremented in the next step 128, the process returns to step 108, affirmative in step 108, or further affirmed in the next step 110, and proceeds to step 130.

  In step 130, after diagnosing a failure of the image forming apparatus 10 based on the classification result that the generated image defect is classified as InOut density unevenness, the process is terminated.

  As described above, in the present embodiment, in the image defect classification process based on the primary test chart, even if the type of the generated image defect cannot be classified into either InOut density unevenness or LoID, the InOut density unevenness And the toner density of the primary test chart is changed (parameters are changed) based on the similarity R of LoID and the secondary test chart determined based on the changed parameters is output, and the image defect caused by the secondary test chart is output. Since the classification process is performed, the generated image defects can be classified with high accuracy into InOut density unevenness. As a result, since the types of image defects can be classified with high accuracy, failure diagnosis can be performed with high accuracy.

[Third Embodiment]
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the present embodiment, as a specific example, a case where it is difficult to classify yellow, magenta, cyan (YMC) simultaneous missing and black (K) single color, which occur in a YMCK solid chart, will be described. Since the present embodiment has substantially the same configuration and processing as the first embodiment, the same parts are denoted by the same reference numerals and detailed description thereof is omitted.

  First, when the image defect diagnosis process (see FIG. 4) is started, in step 100, C (cyan) 30%, M (magenta) 30%, Y (yellow) 30%, K (black) 30 as primary test charts. The print engine unit 14 is instructed to output a test chart printed with being superimposed on the entire surface of the image forming area on the recording medium with a toner density of%. The primary test chart after output is shown in FIG. In the present embodiment, it is assumed that an image defect has occurred on the left side of the primary test chart in the main scanning direction.

  In the next step 102, the image reading unit 16 is instructed to read the output primary test chart, and the process proceeds to step 104. In step 104, an image defect area is detected.

  In the next step 106, image defect classification processing is performed. In this embodiment, since the densities of the R component, G component, and B component of the primary test chart are similar as shown in FIG. 21B, the sum of image defect sizes and image defect areas in the respective color components. Even if the feature values such as the sum of the density values are calculated, there is almost no difference, and it is difficult to classify which color component caused the image defect. That is, in step 200 to step 208 of the image defect classification process shown in FIG. As a result of the calculation, since any value is less than the threshold value, it is determined that it is impossible to specify which color component of the R component, the G component, and the B component has caused the image defect. Therefore, as a result of the image defect classification process, it is determined that it is impossible to specify a color component in which an image defect has occurred.

  Therefore, if a negative determination is made in the next step 108, a negative determination is made in the next step 110, and if a negative determination is made in the next step 112, the process proceeds to step 116.

  In step 116, the parameters of the test chart are changed. The parameter is changed based on a parameter change rule set in advance and stored in the storage unit 22. In the present embodiment, the table shown in FIG. 5 is used as the parameter change rule. In the case of the present embodiment, since the color component in which the defective area has occurred cannot be specified, there are two types of parameters to be changed, parameter 1 and parameter 2, and the parameter is changed to one of the parameters. With parameter 1, the color components of the primary test chart are changed to yellow 30%, magenta 30%, and cyan 30%. In parameter 2, the color component of the primary test chart is changed to 30% black.

  Based on the parameter changed in the next step 118, the secondary test chart is determined. In the next step 120, the print engine unit 14 is instructed to output the determined secondary test chart. A secondary test chart in which the parameters are changed based on the parameter 1 is shown in FIG. FIG. 22B shows the concentrations of the R component, G component, and B component of the secondary test chart. Further, FIG. 23 shows a secondary test chart in which the parameter is changed based on the parameter 2. Further, FIG. 23B shows the concentrations of the R component, G component, and B component of the secondary test chart. In the next step 122, the image reading unit 16 is instructed to read the output secondary test chart, and the process proceeds to step 124.

  In step 124, the output secondary test chart image defect area is detected, and in the next step 126, image defect classification processing is performed. As shown in FIG. 22 and FIG. 23, as a result of the image defect classification process, the defect is identified as a YMC simultaneous missing defect by the secondary test chart.

  Thus, after the loop counter 30 is incremented in the next step 128, the process returns to step 108, affirmative in step 108, or further affirmed in the next step 110, and proceeds to step 130.

  In step 130, after diagnosing the failure of the image forming apparatus 10 based on the classification result that the generated image defect is classified as YMC simultaneous omission, the present process is terminated.

  As described above, in the present embodiment, even if the type of image defect that has occurred cannot be classified as either YMC simultaneous omission or K single color omission in the image defect classification process using the primary test chart, This occurs because the color component of the test chart is changed (parameter is changed), the secondary test chart determined based on the changed parameter is output, and the image defect classification processing is performed by the secondary test chart. Image defects can be classified with high accuracy by simultaneous removal of YMC. As a result, the types of image defects can be classified with high accuracy, so that fault diagnosis can be performed with high accuracy.

[Fourth Embodiment]
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the present embodiment, as a specific example, InOut density unevenness actually occurs, but it is difficult to classify the density as too low (LoID), and yellow, magenta, and cyan that occur in a YMCK solid chart. A case will be described where (YMC) simultaneous omission and black (K) single color omission are difficult to classify simultaneously. Since the present embodiment has substantially the same configuration and processing as the first embodiment, the same parts are denoted by the same reference numerals and detailed description thereof is omitted.

  First, when the image defect diagnosis process (see FIG. 4) is started, in step 100, C (cyan) 30%, M (magenta) 30%, Y (yellow) 30%, K (black) 30 as primary test charts. The print engine unit 14 is instructed to output a test chart printed with being superimposed on the entire surface of the image forming area on the recording medium with a toner density of%. In the next step 102, the image reading unit 16 is instructed to read the output primary test chart, and the process proceeds to step 104. In step 104, an image defect area is detected.

  In the next step 106, image defect classification processing is performed. FIG. 24 is a flowchart illustrating an example of the image defect classification process according to the present embodiment.

  Steps 200 to 208 are the same as the image defect classification process (FIG. 8) of the first embodiment, and thus detailed description thereof is omitted.

  In step 208, it is determined whether or not a color component in which an image defect has occurred is specified. If a negative determination is made, the process proceeds to step 210, and it is determined that the color component cannot be specified (the color component is unknown). Then, the process proceeds to Step 700.

  In step 700, it is determined whether one type of image defect can be specified. The determination of whether or not an image defect can be specified as one type is performed by, for example, performing a process of specifying the type of defect in the test chart (the process from step 212 to step 218) without specifying a color component, and the result. Therefore, it is sufficient to make a judgment.

  If a negative determination is made in step 700, the process proceeds to step 702. In step 702, it is determined that the color component of the generated image defect is unknown and the type of image defect cannot be specified (level 3), and the present process is terminated. On the other hand, if an affirmative determination is made in step 700, the process proceeds to step 704. In step 704, it is determined that the color component of the generated image defect is unknown and the type of the image defect can be specified (level 2), and the present process ends.

  On the other hand, if an affirmative determination is made in step 208, the process proceeds to step 212, and after step 212 to step 218, it is determined in step 220 whether or not an image defect can be specified as one type. If a negative determination is made, the process proceeds to step 706. In step 706, it is determined that the color component of the image defect that has occurred is specified and the type of the image defect cannot be specified (level 1), and this process ends. On the other hand, if an affirmative determination is made in step 220, the process proceeds to step 708, where it is determined that the color component of the generated image defect can be specified and the type of image defect can be specified, and the present process ends.

  Since the subsequent steps 108 to 114 after the completion of the image defect classification processing in step 106 are the same as those in the first embodiment, detailed description thereof is omitted.

  In the next step 116, the parameters of the test chart are changed. The parameter is changed based on a parameter change rule set in advance and stored in the storage unit 22. In the present embodiment, the table shown in FIG. 25 is used as the parameter change rule. Based on the table shown in FIG. 25, the parameters are changed so as to change the color component and toner density corresponding to each level (level 1 to level 3).

  In the case of the present embodiment, since it is not possible to specify either InOut density unevenness or LoID and the color component is unknown, there are two types of parameters to be changed, parameter 1 and parameter 2. Change to any parameter. In parameter 1, the color component of the primary test chart is changed to only YMC, and the toner density of the specified color component is changed to a change amount according to the relationship of FIG. In parameter 2, the color component of the primary test chart is changed to only K, and the toner density of the specified color component is changed to a change amount according to the relationship of FIG.

  Based on the parameter changed in the next step 118, the secondary test chart is determined. In the next step 120, the print engine unit 14 is instructed to output the determined secondary test chart. An image of the output secondary test chart is shown in FIG. In the next step 122, the image reading unit 16 is instructed to read the output secondary test chart, and the process proceeds to step 124.

  In step 124, the output secondary test chart image defect area is detected, and in the next step 126, image defect classification processing is performed. After incrementing the loop counter 30 in the next step 128, the process returns to step 108, affirmative in step 108, or affirmative in the next step 110, and proceeds to step 130. In step 130, after performing a fault diagnosis of the image forming apparatus 10 based on the classification result of the image defect classification process, the present process is terminated.

  As described above, in the present embodiment, in the image defect classification process based on the primary test chart, the level 1 to the level 3 can be classified by the image defect classification process, and the parameter change rule corresponding to each level (see FIG. 25) is stored in the storage unit 22, so that a secondary test chart suitable for diagnosis of an image defect can be output, so that the image defect can be classified with high accuracy. As a result, since the types of image defects can be classified with high accuracy, failure diagnosis can be performed with high accuracy.

  In addition, even when it was difficult to classify into any one of InOut density unevenness and LoID and when it was difficult to classify YMC simultaneous missing and K single color missing at the same time, the parameters of the primary test chart were changed and changed. Since the secondary test chart determined based on the parameters is output and image defect classification processing is performed using the secondary test chart, the generated image defects can be classified with high accuracy. Fault diagnosis can be performed.

[Fifth Embodiment]
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the present embodiment, a configuration in the case of performing image processing and failure diagnosis processing (image defect diagnosis processing including both processes) using an image processing program and a failure diagnosis processing program will be described. Since the present embodiment has substantially the same configuration as that of the first embodiment, the same portions are denoted by the same reference numerals and detailed description thereof is omitted.

  FIG. 26 shows a schematic configuration of an image forming apparatus 10 ′ including the image processing apparatus and the failure diagnosis apparatus according to the present embodiment. The storage unit 22 stores a program (image defect diagnosis processing program) 40 including an image processing program and a failure diagnosis processing program. The program 26 is read from the storage unit 22 and executed by the CPU 26 via the bus 36.

  Note that the image defect diagnosis processing may be performed by recording in a CD-ROM 42, DVD-ROM 44, or other recording medium (not shown) and installing the storage unit 22 via the bus 36.

  In this embodiment, the image defect diagnosis processing program is configured and used as a single program including the image processing program and the failure diagnosis processing program. However, the image processing program and the failure diagnosis processing program are different from each other. You may comprise as a program.

1 is a block diagram illustrating a schematic configuration of an image forming apparatus according to a first embodiment. 1 is a functional block diagram showing a schematic configuration of an image processing apparatus according to a first embodiment. It is a functional block diagram which shows schematic structure of the failure diagnosis apparatus concerning 1st Embodiment. It is a flowchart of the image defect diagnostic process which concerns on 1st Embodiment. It is explanatory drawing which shows the table | surface of an example of the parameter change rule which concerns on 1st Embodiment. FIG. 6 is an explanatory diagram for explaining an example of a relationship between a similarity and a toner density change amount according to the first embodiment. FIG. 10 is an explanatory diagram for explaining another example of the relationship between the similarity and the toner density change amount according to the first embodiment. It is a flowchart of the image defect classification | category process which concerns on 1st Embodiment. It is explanatory drawing for demonstrating the distance of the feature value of RGB each color component in the feature space which concerns on 1st Embodiment, and an origin. It is a flowchart of the image defect color component specific determination process which concerns on 1st Embodiment. It is a flowchart of the image defect kind specific determination process which concerns on 1st Embodiment. It is a flowchart of the similarity calculation process which concerns on 1st Embodiment. It is explanatory drawing for demonstrating the similarity of the kind of image defect which concerns on 1st Embodiment. It is a flowchart of a failure diagnosis process according to the first embodiment. It is explanatory drawing for demonstrating an example of the Bayesian network used by the failure diagnosis process which concerns on 1st Embodiment. It is explanatory drawing which shows the primary test chart which concerns on 2nd Embodiment. It is explanatory drawing which shows the detection result of the image defect area | region of the primary test chart which concerns on 2nd Embodiment. It is explanatory drawing for demonstrating the relationship between the center value vector of the feature value of the defect of the primary test chart which concerns on 2nd Embodiment, and the center value vector of each kind of defect. It is explanatory drawing which shows the secondary test chart which concerns on 2nd Embodiment. It is explanatory drawing for demonstrating the relationship between the center value vector of the feature value of the defect of the secondary test chart which concerns on 2nd Embodiment, and the center value vector of each kind of defect. It is explanatory drawing for demonstrating each component of RGB of the primary test chart and defect area | region which concerns on 3rd Embodiment. It is explanatory drawing for demonstrating each component of RGB of the secondary test chart changed into the parameter 1 which concerns on 3rd Embodiment, and a defect area | region. It is explanatory drawing for demonstrating each component of RGB of the secondary test chart changed into the parameter 2 which concerns on 3rd Embodiment, and a defect area | region. It is a flowchart of the image defect classification | category process which concerns on 4th Embodiment. It is explanatory drawing which shows the table | surface of an example of the parameter change rule which concerns on 4th Embodiment. It is a block diagram which shows schematic structure of the image forming apparatus which concerns on 5th Embodiment. It is explanatory drawing for demonstrating the classification | category of the conventional YMC simultaneous omission and K single color omission. It is explanatory drawing for demonstrating the classification | category of the conventional InOut density nonuniformity and LoID. It is explanatory drawing using the feature space for demonstrating the classification | category of the conventional InOut density nonuniformity and LoID. It is explanatory drawing for demonstrating the classification of the conventional banding and a line / streaks.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Image forming apparatus 14 Print engine part 16 Image reading part 32 Image processing part 34 Fault diagnosis part 40 Program 50 Image processing apparatus 52 Primary test chart output means 54 First image defect area detection means 56 First image defect classification means 58 Similarity calculation means 60 Parameter change means 64 Secondary test chart output means 66 Second image defect area detection means 68 Second image defect classification means 70 Failure diagnosis device 72 Feature quantity extraction unit 74 Feature quantity calculation means 76 Failure identification means

Claims (11)

Primary inspection target image output means for outputting a predetermined primary inspection target image;
First defect area detection means for detecting an image defect area in which an image defect has occurred from image data of an image obtained by reading the primary inspection object image output from the primary inspection object image output means;
First image defect classification means for classifying image defects occurring in the image defect area;
Similarity calculation means for calculating the similarity between each of the generated image defects and a plurality of types of predetermined image defects according to the classification result of the first image defect classification means,
Parameter changing means for changing a parameter relating to a primary inspection target image based on the classification result of the first image defect classification means and the similarity calculated by the similarity calculating means;
Secondary inspection target image output means for outputting a secondary inspection target image based on the parameter changed by the parameter changing means;
Second defect area detection means for detecting an image defect area in which an image defect has occurred from image data of an image obtained by reading the secondary inspection object image output from the secondary inspection object image output means;
Second image defect classification means for classifying image defects occurring in the image defect area of the secondary inspection target image;
An image processing apparatus.   The first image defect classification means classifies color components of image defects and types of image defects occurring in the primary inspection object image, and the second image defect classification means includes the secondary inspection object image. The image processing apparatus according to claim 1, wherein color components of image defects occurring in the image and types of image defects are classified.   The similarity calculation means calculates a feature value of an image defect occurring in the primary inspection target image, and has a relationship between the calculated feature value and a feature value predetermined for each of the plurality of types of image defects. The image processing apparatus according to claim 1, wherein the similarity is calculated based on each.   The image processing apparatus according to claim 3, wherein the feature value includes a size, a density, and a frequency component of the image defect.   The image processing apparatus according to claim 3, wherein the similarity calculation unit calculates a difference between predetermined feature values for each of the plurality of types of image defects.   The image processing apparatus according to claim 1, wherein the parameter changing unit changes at least one of a toner density and a color component of the primary inspection target image. Feature quantity extraction means for extracting feature quantities relating to the degree of image defects occurring in the secondary inspection target image;
Based on the feature amount extracted by the feature amount extraction unit and the classification result of the second image defect classification unit of the image processing apparatus according to any one of claims 1 to 6, Identifying means for identifying the cause of the image defect as the cause of failure,
Fault diagnosis device with   The said specifying means specifies the said failure cause using the Bayesian network selected according to the said classification result from the several Bayesian networks provided for every kind of predetermined image defect. The described failure diagnosis device.   The failure diagnosis apparatus according to claim 7, further comprising a display unit that displays a specifying result of the specifying unit. Outputting a predetermined primary inspection target image;
Detecting an image defect area in which an image defect has occurred from image data of an image obtained by reading the output primary inspection target image;
Classifying image defects occurring in the image defect area;
According to the classification result of the image defect, calculating the similarity between the image defect that has occurred and a plurality of predetermined types of image defects,
Changing a parameter relating to a primary inspection target image based on the classification result of the image defect and the similarity;
Outputting a secondary inspection target image based on the changed parameter;
Detecting an image defect area in which an image defect has occurred from image data of an image obtained by reading the output image of the secondary inspection;
Classifying image defects occurring in an image defect area of the secondary inspection target image;
An image processing program for causing a computer to execute processing including Extracting a feature amount related to the degree of image defects occurring in the secondary inspection target image;
The image defect occurs based on the extracted feature amount and a classification result of the second image defect classification unit of the image processing apparatus according to any one of claims 1 to 6. Identifying the cause as the cause of failure;
A fault diagnosis program for causing a computer to execute processing including