JP4710495B2 - Failure diagnosis system, image forming apparatus, and failure diagnosis method - Google Patents

Description

  The present invention relates to a failure diagnosis system for an image forming apparatus such as a copying machine or a printer, an image forming apparatus, and a failure diagnosis method.

  Conventionally, in office equipment such as copiers and printers, specialized service personnel have been dispatched to perform regular maintenance in order to maintain good quality. However, with the recent trend toward color and high functionality of office equipment, the state of failure has become more complex, and even specialist service personnel cannot fully identify the cause of failure and minimize downtime on the customer side. Therefore, there are many cases in which a plurality of parts that are likely to be related to failure are replaced together. For this reason, there is a problem in that normally normal parts are also replaced together, resulting in an increase in service cost. In order to deal with this problem, a mechanism for automatically diagnosing a failure by using a rule-based system (rule type system) is considered. As an example, there is a failure diagnosis system using a Bayesian network. As a failure diagnosis system using a Bayesian network, for example, there is a technique described in Patent Document 1. According to Patent Document 1, a server for a diagnostic system is provided in a service center or the like, and a customer's printer (image forming apparatus) is diagnosed while exchanging data via the Internet. It introduces the technology to be performed using.

JP 2001-75808 A

  However, in the technology described in Patent Document 1, information necessary for diagnosis is obtained by a method in which a customer directly observes the state of a device or a printed material and inputs it. For this reason, if the customer is not accustomed to observing the malfunction status of the printer, there is a possibility that the input information varies and accurate diagnosis cannot be performed. In addition, since there are many actions to be imposed on the customer before the problem is solved, there is a problem that the user is stressed.

  SUMMARY OF THE INVENTION The present invention has been made in view of such problems, and is capable of accurately acquiring information necessary for failure diagnosis without stressing the user and performing efficient maintenance work. The purpose is to provide a system.

In order to achieve the above object, a failure diagnosis system of the present invention is a failure diagnosis system for diagnosing a failure that occurs in an image forming apparatus, and analyzes a failure diagnosis model that models the cause of the failure of the image forming apparatus. A fault diagnosis unit for diagnosing a fault of each constituent member constituting the image forming apparatus, an internal state information acquisition unit for acquiring internal state information of the apparatus input to the fault diagnosis model, and an image defect that has occurred Selection means for allowing the user to select the type of operation from the operation screen, output means for outputting a test pattern corresponding to the type of image defect selected by the selection means, and optically outputting the test pattern output by the output means analyzing the damage caused to the test pattern by detecting the difference between the inspection image obtained by reading the reference image for inspection That an image defect detecting means, with respect to failure of the test pattern analyzed by the image defect detecting means, the shape, size, density, contour state, defect generation direction of the periodicity, in said selecting means A feature amount extraction unit that extracts a feature amount according to the type of the selected image defect, and the failure diagnosis unit uses the feature amount extracted by the feature amount extraction unit and the internal state information. The cause of the failure is specified (claim 1).

  In the fault diagnosis system of the present invention, (1) shape, (2) size, (3) density, (4) state of contour, (5) direction of occurrence of fault, A feature amount such as a characteristic is extracted, and a cause of failure is specified by a failure diagnosis model in which information on the extracted feature amount and internal state information of the image forming apparatus are input.

  Here, the types of defects commonly found in image forming apparatuses include the generation of lines / bands, the generation of black spots / color spots, the generation of white lines / white bands, white spots, background smudges, and poor color appearance. . Such failure may occur in different ways depending on the cause of failure. Therefore, in the present invention, what kind of trouble is identified, and further, a feature quantity that is considered to be effective for fault diagnosis is extracted for each fault, and the importance of this feature quantity in fault diagnosis is increased. Therefore, it is intended to perform more accurate failure diagnosis.

  For example, when the type of defect is occurrence of a line / band or white line / white band, at least (1) shape information as width information and (2) length information as size information as the feature amount (4) Concentration density information as contour status information, (5) Information regarding whether the paper feed direction is perpendicular to the paper feed direction, or (6) Periodicity information The center of gravity interval information of a plurality of defect areas when there is periodicity and periodicity as information is extracted (claim 3). Further, when the type of the defect is the occurrence of a black point / color point, the feature amount is at least (2) area information as size information, and (3) average density for each color component of the defect region as density information. Information, (6) Existence of periodicity as information on periodicity, and information on the center-of-gravity intervals of a plurality of defective areas when there is periodicity (claim 4). When the type of the defect is blank, the feature amount is at least (2) area information as size information, and (6) a plurality of defects when there is periodicity and periodicity as periodicity information. The center-of-gravity interval information of the region is extracted (claim 5). Further, when the type of the defect is dirt, at least (2) area information as size information, (3) average density information of the defect area as density information, and (6) periodicity Information on the center-of-gravity intervals of a plurality of defective areas in the presence or absence of periodicity and periodicity is extracted as information (claim 6). In addition, when the type of defect is poor color appearance, at least (2) area information as size information and (3) average density information for each color component of the defect area as density information are extracted as the feature amount. (Claim 7).

  In this way, more accurate failure diagnosis can be performed by preferentially inputting information that seems to be more relevant to the failure and performing failure diagnosis. In addition, since the extraction of these feature quantities is performed without the user, it is possible to improve efficiency by eliminating the trouble of inputting the defect information one by one by the user, and a detailed and accurate diagnosis without specialized knowledge about the apparatus. Is possible. In this specification, a user refers to a person who performs an input operation of input information for performing a failure diagnosis.

  If the above-described failure diagnosis system is installed in various image forming apparatuses, the image forming apparatus of the present invention can be configured (claim 8).

The failure diagnosis method of the present invention is a failure diagnosis method for diagnosing a failure occurring in the image forming apparatus, the step of causing the user to select the type of the image defect that has occurred from the operation screen, and the type of the selected image defect. The test pattern is analyzed by detecting a difference between an inspection image obtained by optically reading the output test pattern and an inspection reference image, and outputting a test pattern corresponding to the test pattern. Features according to the type of image defect selected by the user among at least the shape, size, density, contour state, defect occurrence direction, and periodicity with respect to the defect of the analyzed test pattern Extracting the quantity, obtaining the internal state information of the image forming apparatus, and extracting the feature quantity and the internal state information. Analyzed by entering a cause causing come to modeled fault diagnosis model, characterized by a step of diagnosing the failure of individual components constituting the image forming apparatus (claim 9). Such a failure diagnosis method can be implemented, for example, by operating the failure diagnosis system of the present invention.

  According to the present invention, the defect area is automatically detected regardless of the user input, and the feature amount is automatically analyzed according to the defect type so that the failure diagnosis can be performed. This makes it possible to save efficiency and improve efficiency, and to make detailed and accurate diagnosis without specialized knowledge about the apparatus.

  Hereinafter, the best mode for carrying out the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a configuration diagram of an image forming apparatus (hereinafter referred to as “apparatus”) 1 according to the present invention. The apparatus 1 of the present invention includes an image reading unit 101 that reads a document image, a print engine unit 102 that forms and outputs a read image or a print-instructed image, paper passage time, driving current, internal temperature, and humidity. Acquired by the sensor unit 103 including a sensor group for obtaining information on the internal state of the apparatus, the fault diagnosis information input unit 104 for inputting information necessary for fault diagnosis, and the fault diagnosis information input unit 104 A failure diagnosis unit 105 that performs failure diagnosis of the device 1 based on each information is configured. The failure diagnosis unit 105 is a main part of the failure diagnosis system according to the present invention.

  FIG. 2 is a block diagram showing a configuration of the failure diagnosis unit 105. The failure diagnosis unit 105 includes various configurations corresponding to the internal state information acquisition means in the present invention. The component status information acquisition unit 201 that acquires component information indicating the operation status of each component based on the internal status information of the device 1 acquired from the sensor unit 103 as observation data information, and the monitoring result of the usage status of the device 1 as history information The history information acquisition unit 202 to acquire, the environment information in the device 1 directly, or the environment information acquisition unit 203 to acquire the environment information in the device 1 acquired from the sensor unit 103, the image read by the image reading unit 101 The image defect detection unit 204 corresponding to the image defect detection means in the present invention that analyzes the defect of the output image by comparing the (inspected image) with the reference image for inspection, and various analysis results from the image defect detection unit 204 The feature quantity extraction unit 205 corresponding to the feature quantity extraction means in the present invention for extracting feature quantities, and the failure information in a state where the operation conditions differ depending on the user operation. Obtained by the additional operation information acquisition unit 206, each of these units, that is, the component state information acquisition unit 201, the history information acquisition unit 202, the environment information acquisition unit 203, the feature amount extraction unit 205, and the additional operation information acquisition unit 206. A failure probability reasoning unit 207 that calculates the probability of failure based on the information obtained and a diagnosis result notification unit 208 that notifies the user of the diagnosis result.

  Further, the failure probability reasoning unit 207 calculates a probability (failure cause probability) that each cause candidate causing the failure will be a main cause of the failure that has occurred based on each acquired information, and this inference engine. The failure candidate extraction unit 210 narrows down failure cause candidates based on the failure cause probability calculated in 209. Here, a Bayesian network is used for the inference engine 209 that calculates the failure cause probability. This Bayesian network represents a problem area where the causal relationship is complex, so the causal relationship between multiple variables is sequentially connected and expressed as a network with a graph structure, and the dependency relationship between variables is represented by a directed graph. It is a thing. The fault diagnosis model in the present invention is constructed using this Bayesian network.

  Note that the fault diagnosis model, the apparatus 1 and the like of the present invention can use conventional fault diagnosis models, image forming apparatuses, and the like.

FIG. 3 is a diagram conceptually illustrating a configuration example of a Bayesian network in the case of performing a fault diagnosis of an image defect system. As shown, the Bayesian network, the failure cause node ND0 representing the cause of cause image defects, a component state node ND1 indicating the status information of the members constituting the image forming apparatus (components), the history information of the device 1 A history information node ND2 representing the environment information node ND3 representing the surrounding environment information in which the apparatus 1 is installed, an observation state node ND4 representing the state information of the image quality defect, and a user representing the follow-up test result information obtained by the user operation An operation node ND5 and a defect type node ND6 are included.

  The failure cause node ND0 is a node representing a cause causing an image defect, and the probability of this portion is calculated to determine whether or not there is a failure. Each node stores a probability table in which probability data representing the strength of the causal relationship is collected. The initial value of this probability data can be determined using data at the time of past failure occurrence or MTBF (Mean Time Between Failure) of the part.

  The component state node ND1 is a node representing a component state, and is information acquired from the sensor unit 103 that observes the component state. Such information includes information such as component temperature, applied voltage, patch density, and remaining amount of color material (for example, toner).

  The history information node ND2 represents the usage status of the apparatus 1, and uses, for example, a history of the number of prints for each component. This number of prints directly affects the state of the component, such as component wear and deterioration.

  The environment information node ND3 is an ambient environment condition that affects the state of the component, and in this embodiment, the temperature and humidity correspond to this. Temperature and humidity affect the image forming conditions and operating conditions of each component.

  The observation state node ND4 represents the observation state of defects generated in the output image, and is information observed and input by the user. For example, there is information such as the shape, size, density, contour, orientation, position, periodicity, and generation area of the defect.

  The user operation node ND5 is information that causes the device 1 to perform the same process by changing the operation condition, and includes information about the changed operation condition.

  The defect type node ND6 represents the type of image defect and includes information such as lines, dots, white spots, and density unevenness. First, after determining the type of image defect that has occurred and confirming the state of this node, information on other nodes (ND1 to ND5) is appropriately input to make a diagnosis, and the cause of the failure is estimated.

These nodes are connected so as to have a relationship of “cause” → “result”. For example, the relationship between the “failure cause node NDO ” and the “observation state node ND4” is based on the “observation state (concentration is low) based on the“ cause ”indicated by the“ failure cause node ”and the“ observation state node ND4 ”. "Stripes, strips, etc.)" appears. On the other hand, the relationship between the “history information node ND2” and the “ failure cause node NDO ” is “cause” (part deterioration, etc.) based on the “state based on history information (a large number of copies, long operation years, etc.)”. The relationship that occurs is established.

  FIG. 4 shows a specific example of a failure diagnosis model in the failure diagnosis system, and represents a Bayesian network when a black line is generated in a configuration example of failure diagnosis due to image defects. As shown in the figure, the nodes are connected so as to have a relationship of “cause” → “result”. For example, the relationship between “drum flaws” and “line width information” is such that “line width information” such as the occurrence of thin lines based on “drum flaws” appears.

  On the other hand, the relationship between “feed number history information” and “fuser” is based on the state based on “number of feeds” (number of feeds or more), and the possibility of black line generation due to “fuser” deterioration increases. It holds.

  The initial value of the probability data of each node is determined based on, for example, past data. Thereafter, the probability of each node may be periodically updated based on market trouble statistical data such as the replacement frequency of parts and the frequency of occurrence of defects. Also, nodes representing image defect features such as “line width information”, “periodic information”, and “occurrence location information” in FIG. 4 are based on the feature amounts obtained by the feature amount extraction unit 205 in FIG. The state is determined.

  Next, an outline of a processing procedure for failure diagnosis in relation to an image defect will be described with reference to a flowchart shown in FIG. First, when the mode is shifted to the failure diagnosis mode, the type of image defect to be subjected to failure diagnosis is displayed on the operation screen. The user selects the type of defect that has occurred on the operation screen and outputs a corresponding test pattern (S501). As the test pattern output at this time, for example, when the type of the selected defect is “line / band”, “black spot / color point”, or “background stain”, a blank test pattern is output. In the case of “white line / white belt”, “white blank”, and “color defect”, a total of four charts of 100% of the entire color of each YMCK are output. In addition, depending on the structure of the apparatus, when a pattern that can easily reproduce a defect is conceivable, such a unique pattern may be used. The test pattern output here is stored in advance in the print engine unit 102 shown in FIG. 1. If the cause of the failure is a component of the print engine unit 102, the defect is reproduced in the test pattern. However, when there is a cause in a part of the image reading unit 101 such as a defect only at the time of copying, the defect is not reproduced in the test pattern. However, when there is a cause in the components of the image reading unit 101, when the test pattern is set in the image reading unit 101 and the output image is read, a defect appears in the read image. Therefore, before reading the output image, an inquiry is made from the operation screen as to whether or not a defect occurs only at the time of copying, and the user can select and input the information. The selected information is acquired from the additional operation information acquisition unit 206 and input to the failure probability inference unit 207. When the test pattern is discharged from the print engine unit 102 of the apparatus 1, the test pattern is set in the image reading unit 101 and the output image is read (S502).

  Next, the image defect detection unit 204 included in the failure diagnosis unit 105 compares the read image with a reference image stored in advance in the apparatus to check whether there is an image defect (S503). If no defect is detected by this process, the previous defect may have been accidental, or it may have been resolved by some action before outputting the test pattern. Is notified to the user through the operation screen and the process is terminated (S504-N). On the other hand, when a defect is detected (S504-Y), the feature amount extraction unit 205 analyzes the state of the detected defect and extracts a feature amount (S505). Here, feature quantities such as (1) shape, (2) size, (3) density, (4) contour state, (5) defect occurrence direction, and (6) periodicity are extracted according to the type of defect. To do. Furthermore, various data necessary for failure diagnosis such as status information of each component constituting the device 1, history information such as a counter value indicating the number of printed sheets for each component, environmental information such as temperature and humidity inside the device, Acquired by the component state information acquisition unit 201, history information acquisition unit 202, and environment information acquisition unit 203 (S506).

  When the failure probability inference unit 207 receives data from the feature amount extraction unit 205 and the various information acquisition units 201 to 203, the failure probability inference unit 207 calculates the occurrence probability of each failure cause using the inference engine 209 corresponding to the detected defect (S507). . Based on the calculated probabilities, the failure candidate extraction unit 210 extracts failure causes for the number of candidates specified from the higher probability of causing a failure (S508). Note that the number of candidates may be set in advance, or an arbitrary number may be input and designated before candidate extraction. The extracted failure cause candidates are displayed on a display device such as a control panel by the diagnosis result notification unit 208 and notified to the user (S509). If the failure cause candidates can be narrowed down at this stage, the process ends (S510-N). However, in such automatic determination processing, failure cause candidates are not necessarily narrowed down to one at this point. Therefore, when failure cause candidates cannot be narrowed down at this point, additional operation items necessary for further failure diagnosis are selected from the operation screen, and the operation conditions of the apparatus 1 are changed according to the selected items, and the image is re-output. (S501-Y). Then, the user inputs information on the follow-up test result from the operation screen (S511). The additional operation at this time is enlargement / reduction of an image, output of a test pattern held at each position of the image path, or the like, and examines whether or not a defect occurrence state has changed. Therefore, the follow-up test result is of a level that allows the user to easily input according to the question on the operation screen. Then, the failure cause probability is recalculated by combining the added information and the previously input information, and failure candidates are narrowed down based on the result. When failure candidates are narrowed down or when there is no information to be added even if narrowing down is not possible, the process ends (S510-N).

  Next, the detailed procedure of the image defect detection process described in S503 in the figure will be described with reference to the flowchart shown in FIG. First, the image read by the image reading unit 101 is divided into RGB components (S601). If the output test pattern is a blank chart, defect detection processing is performed in the order of R component → G component → B component. On the other hand, in the case of a YMCK 100% chart, only the complementary B component for the Y chart, only the complementary G component for the M chart, only the complementary R component for the C chart, and only for the K chart Since all components have the same output, only the G component is used, for example. If test patterns are output in the order of Y → M → C → K, defect detection processing is performed in the order of B component of the Y test pattern → G component of the M test pattern → R component of the C test pattern → G component of the K test pattern. I do.

  In this defect detection process, first, a binarization process is performed on a first component image using a predetermined threshold (S602). Next, in order to remove the noise component and extract only the defect area, a morphology detection process is performed by applying a defect detection filter (S603). An example of the filter used at this time is shown in FIG. 7A shows a horizontal line detection filter, FIG. 7B shows a vertical line detection filter, and FIG. 7C shows a point or irregular defect detection filter. When “line / band” or “white line / white band” is selected on the operation screen by the user, noise component removal and defect area detection are first performed using a horizontal line detection filter. If a defective area is detected at this time, the morphology process is terminated. On the other hand, if a defective area is not detected, noise component removal and defective area detection are performed using a vertical line detection filter. If the selected defect type is other than that, the filter of FIG. 7C is used. When extraction processing for defect detection is performed in this way, it is determined whether there is a color component to be further processed (S604). Here, when there is a color component to be further processed (S604-Y), the color component images are sequentially read in the order described above (S605), and the processing of S602 to S603 is repeated. When the processing for all component images is completed (S604-N), the process proceeds to S503 shown in FIG. If the device that performs failure diagnosis is a black and white machine, the detection process may be performed using only a single grayscale image.

  Next, details regarding the defect feature amount extraction processing described in step 505 of FIG. 5 will be described using the flowchart shown in FIG. Here, the feature quantity to be extracted differs depending on the type of defect. The feature quantity extracted for each defect type is as follows.

  First, if the defect is “line / band” or “white line / white band”, depending on the cause of failure, the direction and thickness / length of the line / band, blurring of the line outline, In the case of the generation direction and the horizontal line, whether or not it occurs at a constant cycle is different. Therefore, the six items (1) shape, (2) size, (3) density, (4) contour state, (5) defect occurrence direction, and (6) periodicity described in S505 of FIG. Among them, (1) line width as shape information, (2) length of line / band as size information, (4) density average of boundary region pixels of defect area as state information of contour, (5) defect information The generation direction, and (6) the center-of-gravity coordinate interval between defect areas as information on the presence or absence of periodicity are extracted as feature quantities.

  When the defect is a “black point / color point”, the size and density of the point, the presence or absence of periodicity, and the interval are different depending on the cause of the failure. Therefore, among the six items described in S505 of FIG. 5, (2) the area of the defect area as size information, (3) the average density for each color component in the defect area as density information, and (6) periodicity Whether or not there are a plurality of defective areas is extracted as feature information.

  When the defect is “white spot”, it has the same characteristics as “black spot / color spot” depending on the cause of the failure. However, if the defect is blank, density information is not necessary, so out of the six items described in S505 of FIG. 5, (2) the area of the defect region as size information, and (6) the period Whether or not there are a plurality of defect areas as sex information, and if so, the center-of-gravity coordinate interval of each area is extracted as a feature amount.

  When the defect is “soil dirt”, the dirt generation range and the dirt density differ depending on the cause of the failure. In addition, dirt may occur with periodicity. Accordingly, among the six items described in S505 of FIG. 5, (2) the size, that is, the area of the defect region as the generation range information, (3) the average density for each color component in the defect region as the density information, ( 6) As a periodicity information, whether or not there are a plurality of defective areas, and if there are, the center-of-gravity coordinate interval of each area is extracted as a feature amount.

  When the defect is “color defect”, the density balance of the output image is locally generated or the density variation is entirely generated due to the cause of the failure. Therefore, among the six items described in S505 of FIG. 5, (2) size, that is, the area of the defect region as information on the generation range, and (3) average density for each color component in the defect region as density information. Extracted as feature quantity.

  Next, a processing flow (FIG. 8) of the feature amount extraction processing will be described. First, the defect areas detected in S503 in the flowchart shown in FIG. 5 are labeled to classify the areas (S801). Next, a circumscribed rectangle of each area is extracted (S802). Then, the width, height, area, and barycentric coordinates of each circumscribed rectangle are calculated (S803). Image quality defects in image forming devices such as copiers and printers do not occur in the diagonal direction for lines and belts, and for irregular shapes such as white spots and black spots, the size is small, so each parameter calculated from the circumscribed rectangle Is not significantly different from the parameters of the actual defect area. From these values, it is possible to obtain feature quantities relating to (1) the shape of the defect and (2) the size of the defect. Regarding (1), if the type of defect that occurred is "line / band" or "white line / white band", the line width can be determined from the width or height of the circumscribed rectangle, whether it is a vertical line or a horizontal line. It is possible to distinguish between “linear” and “strip” using a predetermined threshold value. Regarding (2), if the type of defect that occurred is "line / band" or "white line / white band", know the length of the line from the width or height of the circumscribed rectangle, whether it is a vertical line or a horizontal line. Can do. If the type of defect that occurred is “black spot / color point”, “white spot”, “background stain”, or “color defect”, the size and occurrence range can be determined from the area of the circumscribed rectangle of the defect area. . Regarding the defect generation direction (5), when the type of the generated defect is “line / band” or “white line / white band”, the type of defect detection filter used in S603 of the flowchart shown in FIG. It is possible to know in which direction the paper feed direction or the direction perpendicular to the paper feed direction has occurred.

  Next, the defect area detected in S503 in the flowchart shown in FIG. 5 is applied to the multivalued image read by the image reading unit 101 as a mask area, and together with the feature amount calculated in S803, (3) density , (4) The feature amount of the defect image relating to the contour state and (6) periodicity is extracted (S804 to S806). With regard to (3), when the type of the generated defect is “black spot / color point”, “background stain”, or “color defect”, the defect detected in S503 of the flowchart shown in FIG. The area is masked, and the average density inside the masked area is calculated (S804).

  Then, the average density calculated by the following method is evaluated to determine the defect state (S807). For example, the average density is evaluated as follows. As shown in FIG. 9A, first, the average density distribution for each defect area is examined, and the average value and variance are calculated. Then, as shown in FIG. 9B, using the relationship diagram of the average value vs. dispersion created based on the actual data, it is determined whether the density of the defect detected this time is close to the data of which failure group. (Evaluation of average concentration) is performed. For example, the Mahalanobis distance may be used for the density determination. Next, with regard to (4), when the type of defect that has occurred is “line / band” or “white line / white band”, pixels at the boundary are extracted from the defect area detected in S503 in FIG. Is calculated (S805). Next, the density of the average density calculated using the threshold value created based on the actual data is discriminated, and it is determined whether the outline is clear / blurred and the state of the defect is determined (S807). When there are a plurality of defective areas, the average of the whole may be obtained, or may be determined using a method similar to the above-described density evaluation method. Next, regarding (6), the periodicity appears when the rotating parts are dirty or scratched, and the direction in which the period appears is with respect to the paper feed direction. Therefore, if the type of defect that has occurred is "line / band", "white line / white band", "black spot / color point", "white spot", or "background stain", the periodicity evaluation is calculated in S803. This is performed depending on whether the interval between the coordinates of the paper feed direction component among the barycentric coordinates of each defective area is constant (S806). If the interval is constant, it is determined which part cycle corresponds to the reference feature amount, and if the interval between coordinates is indefinite, it is determined that there is no periodicity (S807).

  If the feature amount can be extracted as described above, the process proceeds to S506 in the flowchart shown in FIG. 5 and is input to the inference engine 209 together with other acquired information to perform failure diagnosis and extract a failure cause candidate. That is, failure diagnosis is performed by inputting information based on the feature quantity to each node of the Bayesian network shown in FIG. 4, that is, the failure diagnosis model.

  Note that, depending on the configuration of the image forming apparatus, the feature amount to be extracted may vary with respect to the above-described defect type. Therefore, a feature amount specific to the device may be additionally extracted and diagnosed.

  The above-described embodiments are merely examples for carrying out the present invention, and the present invention is not limited to these embodiments. Various modifications of these embodiments are within the scope of the present invention. It is obvious from the above description that various other embodiments are possible within the scope of the above.

1 is a block configuration diagram of an image forming apparatus according to an embodiment. It is a block diagram which shows the example of 1 structure of the failure diagnosis part concerning this invention. It is an example of composition of a Bayesian network in the case of performing a fault diagnosis of an image defect system concerning the present invention. It is a Bayesian network at the time of black line generation in the structural example of the failure diagnosis by the image defect concerning this invention. It is a flowchart which shows the process sequence of the failure diagnosis system concerning this invention. It is a flowchart which shows the detection process procedure of an image defect. It is a figure which shows the example of the filter for defect detection. It is a flowchart which shows the feature-value extraction processing procedure of the image defect concerning this invention. It is a figure explaining the density determination method of defect density.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image forming apparatus 101 Image reading part 102 Print engine part 103 Sensor part 104 Failure diagnosis information input part 105 Failure diagnosis part 201 Component state information acquisition part 202 History information acquisition part 203 Environmental information acquisition part 204 Image defect detection part 205 Feature amount extraction Unit 206 additional operation information acquisition unit 207 failure probability inference unit 208 diagnosis result notification unit 209 inference engine 210 failure candidate extraction unit

Claims (9)

A failure diagnosis system for diagnosing a failure occurring in an image forming apparatus,
A failure diagnosis unit for diagnosing a failure of each component constituting the image forming apparatus by analyzing a failure diagnosis model that models a cause of failure of the image forming apparatus;
Internal state information acquisition means for acquiring internal state information of the device input to the fault diagnosis model;
Selection means for allowing the user to select the type of image defect that has occurred from the operation screen;
Output means for outputting a test pattern corresponding to the type of image defect selected by the selection means;
An image defect detecting means for analyzing a defect occurring in the test pattern by detecting a difference between an inspection image obtained by optically reading the test pattern output by the output means and a reference image for inspection; ,
The type of the image defect selected by the selection means among the shape, size, density, contour state, failure occurrence direction, and periodicity for the test pattern defect analyzed by the image defect detection means And feature amount extraction means for extracting feature amounts according to
The failure diagnosis unit is characterized by specifying a cause of a failure using the feature amount extracted by the feature amount extraction unit and the internal state information.   The type of defect analyzed by the image defect detection means is any of generation of a line / band, generation of a black / color point, generation of a white / white band, white spot, background stain, and poor color appearance. The fault diagnosis system according to claim 1, wherein:   When the type of defect is occurrence of a line / band or white line / white band, at least the shape information as width information, the length information as size information, and the outline portion as outline state information Density information, information on whether the direction of occurrence of the defect is the paper feed direction or the direction perpendicular to the paper feed direction, presence / absence of periodicity as periodicity information, and multiple defect areas with periodicity 3. The fault diagnosis system according to claim 2, wherein the center-of-gravity interval information is extracted.   When the type of defect is black point / color point occurrence, area information as at least size information as the feature amount, average density information for each color component of the defect region as density information, and period as periodicity information The fault diagnosis system according to claim 2, wherein the center-of-gravity interval information of a plurality of defect areas in the presence or absence of periodicity and periodicity is extracted.   When the type of defect is blank, area information as at least magnitude information as the feature amount, presence / absence of periodicity as periodicity information, and center-of-gravity interval information of a plurality of defective areas when periodicity exists The fault diagnosis system according to claim 2, wherein the fault diagnosis system is extracted.   In the case where the type of the defect is dirt, there is area information as at least the size information as the feature amount, average density information of the defect area as density information, presence / absence of periodicity and periodicity as periodicity information The failure diagnosis system according to claim 2, wherein the center-of-gravity interval information of a plurality of defective areas is extracted.   2. The method according to claim 1, wherein when the defect type is poor color appearance, area information is extracted as at least size information as the feature amount, and average density information for each color component of the defect area is extracted as density information. 2. The fault diagnosis system according to 2. An image forming apparatus comprising the failure diagnosis system according to claim 1. A failure diagnosis method for diagnosing a failure occurring in an image forming apparatus,
A step of allowing the user to select the type of image defect that has occurred from the operation screen;
Outputting a test pattern corresponding to the type of the selected image defect;
Analyzing the defect of the test pattern by detecting a difference between the inspection image obtained by optically reading the output test pattern and a reference image for inspection;
A feature amount corresponding to the type of image defect selected by the user is extracted from at least the shape, size, density, contour state, defect occurrence direction, and periodicity for the analyzed test pattern defect. And a process of
Obtaining internal state information of the image forming apparatus;
The extracted feature quantity and the internal state information are input and analyzed in a failure diagnosis model that models the cause of the failure of the image forming apparatus, and the failure of the individual components constituting the image forming apparatus is diagnosed. A failure diagnosis method comprising the steps of:

Images