Inspection device and control method for inspection device

Abstract

To achieve simple inspection setting and improve inspection accuracy by detecting a factor configuring an image regardless of a form of receiving a manuscript and setting an inspection method according to the detected factor.SOLUTION: An inspection device of the present invention includes: registration means that registers a received image as a reference image; detection means that analyzes the image to detect a factor configuring the image; setting means that sets an inspection condition including at least an inspection method on the basis of the detected factor; acquisition means that acquires a read image by reading a printed matter obtained by printing the image on a recording sheet; and inspection means that inspects the printed matter under the inspection condition set by the setting means on the basis of the read image and the reference image.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The present invention relates to an inspection apparatus and a control method for the inspection apparatus. [Background technology]

[0002] There is known an inspection device that reads a printed matter printed by a printing device and detects defects such as stains and missing prints on the printed matter. In order to perform the inspection, it is necessary to set an inspection area in the area to be inspected on the image and set an inspection level for detecting defects in the inspection area. However, manually setting appropriate inspection settings is time-consuming in the inspection process. In order to solve such problems of inspection settings, Patent Document 1 proposes a technology that analyzes internal data of PDF, detects images, graphics, and characters, and sets an inspection area and an inspection level according to each object. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-107788 Summary of the Invention [Problem to be solved by the invention]

[0004] However, in Patent Document 1, since information inside the PDF data is analyzed to obtain object information, which is an element that constitutes an image, it is not possible to obtain object information when data other than the PDF format is submitted.Furthermore, there is a problem in that it is not possible to set different inspection methods for images, graphics, and characters.

[0005] The present invention aims to realize simple inspection settings and improve inspection accuracy by detecting the elements that make up an image regardless of the submission format and then setting an inspection method according to the detected elements. [Means for solving the problem]

[0006] In order to achieve the above-mentioned object, the inspection device of the present invention is characterized by having a registration means for registering a received image as a reference image, a detection means for analyzing the image and detecting elements that constitute the image, a setting means for setting inspection conditions including at least an inspection method based on the detected elements, an acquisition means for acquiring a read image by reading a printed matter on which the image is printed on a recording sheet, and an inspection means for inspecting the printed matter according to the inspection conditions set by the setting means based on the read image and the reference image. Effect of the Invention

[0007] According to the present invention, inspection settings including an inspection method can be automatically set regardless of the format of input data, thereby improving inspection accuracy. [Brief description of the drawings]

[0008] [Figure 1] 1 is a diagram showing an example of the overall configuration of an image forming apparatus 1 according to the present embodiment. [Diagram 2] FIG. 2 is a diagram illustrating image data in the present embodiment. [Diagram 3] FIG. 2 is a diagram illustrating the flow of various data in the present embodiment. [Figure 4] 4 is a flowchart illustrating a control flow in the present embodiment. [Diagram 5] FIG. 4 is a diagram illustrating a setting screen according to the present embodiment. [Figure 6] FIG. 2 is a diagram illustrating each region in the present embodiment. [Figure 7] FIG. 2 is a diagram illustrating an inspection means in the present embodiment. [Figure 8] 13 is a diagram illustrating a combined image of Nup in another embodiment to which this embodiment is applied. FIG. [Figure 9] FIG. 11 is a diagram illustrating the flow of various data in still another embodiment to which the present embodiment is applied. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, the embodiments of the present invention will be described in detail with reference to the accompanying drawings. Note that the following embodiments do not limit the scope of the present invention, and not all of the combinations of features described in the embodiments are necessarily essential to the solution of the present invention.

[0010] (First embodiment below) Hereinafter, an image forming apparatus according to the present embodiment will be described with reference to the drawings.

[0011] FIG. 1 is a diagram showing an example of the overall configuration of an image forming apparatus 1000 in the first embodiment to which the present embodiment is applied. As shown in FIG. 1, the image forming apparatus 1000 includes an image forming apparatus main body 1100, an image reading device 1300, a paper discharge unit 1400, and an inspection PC 1500. A client PC 1200, which is an example of a terminal device, is connected to the image forming apparatus main body 1100 via a communication line 1201 such as a network. The client PC 1200 transmits print data to the image forming apparatus main body 1100. The image forming apparatus main body 1100 receives the print data transmitted by the client PC 1200 via the communication line 1201, and performs printing based on the received print data. A reading unit 1110 is provided on the upper part of the image forming apparatus main body 1100. The reading unit 1110 includes an ADF 1110A and a document reading unit 1110B. The ADF 1110A includes an original tray 1111, a paper path 1112, a paper discharge tray 1113, a contact image sensor 1114, and a density reference member 1115. The density reference member 1115 is used when performing shading correction for the ADF 1110A. The original reading unit 1110B includes an original illumination unit 1116, a reflecting mirror 1117, a condenser lens 1118, a sensor 1119, and a platen glass 1120. The reading unit 1110 separates and feeds out the originals set in the original tray 1111 one by one, transports them in the sub-scanning direction along a paper path 1112 on which the contact image sensor 1114 is disposed, and discharges them to the paper discharge tray 1113. The original illumination unit 1116 includes a lamp 1116A and a mirror 1116B. While the document is transported in the sub-scanning direction along a paper path 1112 , a reading operation is repeatedly performed line by line in the main scanning direction by a document illumination unit 1116 , a reflecting mirror 1117 , a condenser lens 1118 , and a sensor 1119 .

[0012] The image forming apparatus main body 1100 includes a control unit 1160, an image forming section 1130, a fixing section 1140, and a paper feed section 1150. The control unit 1160 performs image processing on the document image data acquired from the client PC 1200 to generate a print image, and transmits the image to the image forming section 1130. Details will be described later. The control unit 1160 also controls the image forming section 1130, the fixing section 1140, and the paper feed section 1150. The image forming section 1130 includes an exposure device 1131, a developing device 1132, a photosensitive drum 1133, and a transfer belt 1134. The image forming section 1130 develops the image by supplying different color toners to the photosensitive drum 1133 by the exposure device 1131 based on the print image data received from the control unit 1160 or the image data of the document read by the reading section 1110. The image forming unit 1130 transfers the toner image developed on the photosensitive drum 1133 by the transfer belt 1134 to the paper supplied from the paper feed unit 1150. The image forming unit 1130 fuses the toner of the toner image transferred to the paper in the fixing unit 1140, thereby fixing a color image to the paper. The image reading device 1300 is disposed in an inline manner at the rear of the image forming apparatus main body 1100, and reads an image printed on one or both sides of the paper on which the image is formed. Note that the following description is based on the image reading device 1300 disposed in an inline manner, but the image reading device 1300 may be disposed in an offline manner at the rear of the image forming apparatus main body 1100.

[0013] The image reading device 1300 includes a control unit 1360, an image reading section 1340A, an image reading section 1340B, a colorimeter 1350, background members 1330A to 1330C, a conveying section 1320, and a conveying path 1310. The control unit 1360 transmits a read image acquired from the image reading section 1340 to the control unit 1160. Details will be described later. The control unit 1360 also controls the conveying section 1320, the background member 1330, the image reading section 1340, the colorimeter 1350, and the like. The conveying path 1310 is a path through which a sheet passes. The conveying section 1320 conveys a sheet on which an image is formed. Thus, the sheet is conveyed along the conveying path 1310 by the driving of the conveying section 1320. For example, when the image reading device 1300 receives a sheet of paper supplied from the image forming apparatus main body 1100, the image reading device 1300 causes the image reading units 1340A and 1340B and the colorimeter 1350 to read the image formed on the sheet. The image reading result (hereinafter, described as a read image) may be output to the image forming apparatus main body 1100 or the like. Specifically, the image reading unit 1340A and the image reading unit 1340B are each disposed at a position facing either the front or back of the sheet of paper passing through the conveying path 1310. The image reading unit 1340A is disposed at a position to read the back side of the sheet of paper. The reading result of the image reading unit 1340A may be used, for example, to check the misalignment between the front and back sides of the image printed on the sheet of paper, or to check the presence or absence of an unexpected image. On the other hand, the image reading unit 1340B is disposed at a position to read the front side of the sheet of paper. Specifically, the image reading unit 1340B reads the image printed on the sheet of paper. As the paper is transported, the image reading unit 1340B reads the color of the image formed on the paper in a direction perpendicular to the paper travel direction, i.e., the main scanning direction. Note that the image reading units 1340A and 1340B are collectively referred to as the image reading unit 1340. The image reading unit 1340 is composed of, for example, a scanner. The background members 1330A to 1330C are collectively referred to as the background member 1330.

[0014] The paper discharge unit 1400 has paper discharge destinations 1410 and 1420, and transport paths 1430, 1440, 1450, and 1460. The paper that has passed through the image reading unit passes through transport path 1430, and is transported from transport path 1440 when paper discharge destination 1410 is used, and from transport path 1450 when paper discharge destination 1420 is used. The inversion unit 1460 inverts the paper once so that the orientation of the incoming paper is the same as the orientation of the paper when it is discharged. The paper discharge destination 1410 may be used when the inspection is OK, and the paper discharge destination 1420 may be used when the inspection is NG.

[0015] The inspection PC 1500 is connected to the image forming apparatus main body 1100 via a communication line 1501 such as a network, and is connected to the image reading device 1502 via a communication line 1502. The inspection PC 1500 receives a print image from the image forming apparatus main body via the communication line 1501, and receives a read image from the image reading device 1300 via the communication line 1502. The received print image is registered as a reference image in the inspection PC 1500. The inspection PC 1500 inspects the finish of the print by comparing the reference image with the read image.

[0016] Next, the region of the inspection image will be described with reference to FIG. 2. FIG. 2 shows print data transmitted from the client PC 1200 to the image forming apparatus main body 1100. As the data structure of the print data, for example, data configured in a page description language (PDL) such as PDF may be used. Also, a raster image configured of pixels and their brightness values ​​such as a bitmap image or PNG (Portable Network Graphics) may be used. In this embodiment, the print data 2001 is configured of an image region object 2002, a character region object 2003, and a background region object 2004. Furthermore, the image region 2002 includes a person object 2005 and a car object 2006.

[0017] Next, the control for inspecting the finish of a printed matter will be described with reference to FIG.

[0018] The control unit 1160 of the image forming apparatus main body 1100 includes a control unit 3001 such as a CPU (Central Processing Unit) and a memory 3002 such as a ROM (Read Only Memory) or a RAM (Rewritable Memory). The image forming apparatus 1100 further includes a storage unit 3003 such as an HDD (Hard Disk Drive), LAN interfaces 3004 and 3005 as communication interfaces, and a video interface 3006 for transferring image data to the image forming apparatus. The control unit 3001 reads out a program from the memory in accordance with the processing contents, executes the program, and controls the operation of the image forming apparatus main body 1100. Various data stored in the storage unit 3003 are referenced when the control unit 3001 controls the operation of the image forming apparatus main body 1100.

[0019] The LAN interface 3004 is a communication interface for receiving print data and the like from the client PC 1200. The LAN interface 3005 is a communication interface for receiving images from the image reading device 1300.

[0020] The control unit 3001 performs screen processing, i.e., raster image processing for halftone dot conversion, so that print data generated by the client PC 1200 can be printed by the image forming apparatus main body 1100. Raster image processing, i.e., RIP processing, generates a raster image. A raster image is data of a rasterized image, which is image data constituting a so-called RIP image. The generated raster image is stored in the storage unit 3003. When printing is performed, the control unit 3001 reads out the raster image stored in the storage unit 3003 and performs further image correction. When image processing is performed, the control unit 3001 performs image processing, generates an image for printing, and stores it in the storage unit 3003. The control unit 3001 also transmits the image for printing to the image forming unit 1130 as a video signal, and the image forming unit 1130 forms the image on paper.

[0021] The image formed on the paper is read by the image reading device 1300. The control unit 1360 of the image reading device 1300 includes a control unit 3101 such as a CPU (Central Processing Unit) and a memory 3102 such as a ROM (Read Only Memory) or a RAM (Rewritable Memory). The control unit 1360 further includes a storage unit 3103 such as a HDD (Hard Disk Drive), a LAN interface 3104 as a communication interface, and an I / O interface 3105 for acquiring read image data acquired by the image reading unit 1340. The control unit 3101 reads a program from the memory 3102 according to the processing contents, executes the program, and controls the operation of the image reading device 1300. Various data stored in the storage unit 3103 are referred to when an example of a control unit such as the control unit 3101 controls the operation of the image reading device 1300.

[0022] The scanned image is transmitted to the control unit 1160 via the control unit 3101 and the LAN interface 3104. The control unit 3001 of the control unit 1160 that has received the scanned image compares the raster image with the scanned image and judges whether or not there is an abnormality in the printed matter, thereby inspecting the finish of the printed matter. Details will be described later.

[0023] The inspection PC 1500 includes a control unit 3201 such as a CPU (Central Processing Unit) and a memory 3202 such as a ROM (Read Only Memory) or a RAM (Rewritable Memory). The inspection PC 1500 also includes a storage means 3203 such as an HDD (Hard Disk Drive) and a LAN interface 3204 as a communication interface. The control unit 3201 reads out a program from the memory according to the processing contents, executes the program, and controls the operation of the inspection PC. Various data stored in the storage unit are referenced when the control unit controls the operation of the inspection PC 1500. The GPU (Graphics Processing Unit) 3205 is a processor that performs calculations required for image processing, and executes commands related to image processing from the control unit 3201.

[0024] Next, a flow of inspecting the finish of a printed matter in the first embodiment will be described with reference to the flowchart of Fig. 4. In this embodiment, a case will be described in which print data 2001 in Fig. 2 is input from client PC 1200 to image forming apparatus main body 1100, and a printed matter on which the print data is printed on a recording sheet is inspected. Note that steps S4002 to S4006 in the flowchart are realized by control unit 3001 such as a CPU reading and executing a program stored in memory 3002. Steps S4102 to S4103 in the flowchart are realized by control unit 3101 such as a CPU reading a program stored in memory 3102. Steps S4202 to S4216 in the flowchart are realized by control unit 3201 such as a CPU of the inspection PC calling a program stored in memory 3202.

[0025] This flowchart starts, for example, when the image forming apparatus main body 1100 is powered on, and proceeds to step S4002. If the control unit 3001 receives print data from the client PC 1200 in step S4002, the process proceeds to step S4003, and if not, the process proceeds to S4002.

[0026] In step S4003, the control unit 3001 performs raster image processing on the received print data to generate a raster image, and stores the raster image in the storage unit 3003. For example, print data such as PDL is rasterized via a display list to generate a raster image.

[0027] In step S4004, a print image is generated from the raster image generated in step S4003 based on print setting information. The print setting information here includes information related to the finish of the printout, such as magnification, Nup, color adjustment, and page number addition.

[0028] In step S 4005 , the control unit 3001 sends the image to be printed to the inspection PC 1500 .

[0029] FIG. 4C is a flowchart of the control unit 3201 of the inspection PC 1500, which starts with receiving an image to be printed from the image forming apparatus main body 1100.

[0030] If the control unit 3201 receives an image to be printed in step S4202, the process proceeds to step S4203.

[0031] In step S4203, the control unit 3201 performs image analysis on the received print image. Specifically, the layout of the print image is analyzed to obtain objects constituting the print image. Here, the object refers to an element constituting an image, and in this embodiment, it has at least an image region object, a character region object, and a background region object. The background region object refers to an area of ​​the entire print image excluding the image region object and the character region object. In addition, a machine learning model trained to analyze the layout stored in the storage unit 3203 in advance is used for layout analysis of the print image. For example, PubLayNet (2021, Zejiang Shen et al.) is an example of a machine learning model trained to analyze the layout. The method of analyzing the layout is not limited to the machine learning model, and characters and images may be detected by analyzing the histogram of brightness values. In this embodiment, an image region object 2002, a character region object 2003, and a background region object 2004 are detected by layout analysis.

[0032] Furthermore, the image area object can be analyzed to detect an object of an element of interest that the user wants to inspect with emphasis, which is included in the image area object. A method of designating an object of an element of interest will be described with reference to FIG. 5. A screen 5000 is a screen displayed on an operation unit (not shown) such as a client PC or an inspection PC, and receives instructions from the user. An object of an element of interest can be designated from an element of interest field 5001 on a screen 5000 for setting print settings. For example, the designation method may be such that an arbitrary element can be selected from options prepared in advance as in 5100. The object of interest can also be selected by accepting input of a word from the user as in 5200. Any number and type of object of an element of interest can be selected. The method of designating an object of an element of interest is not limited to these, and for example, a type of deliverable can be selected, and when a poster or photo book is selected, a selection screen of an object of an element of interest can be displayed. On the other hand, when an image area object is not considered important, such as a form, a selection screen of an object of an element of interest can be switched so that it is not displayed. In addition, an image recognition model can be used to automatically determine an element of interest. When an element designated in this way is included in an image area, inspection conditions are also set for that area. For object detection in an image region, a machine learning model trained to detect objects in an image is used. For example, Detectron2 (Facebook AI Research) can be mentioned. In this embodiment, a case where a "car" is selected as an object of interest element will be described. In FIG. 2, a car object 2006 and a person object 2005 are detected in an image region object 2002, but the car object 2006 becomes an object of interest element, and the person object 2005 is treated as an image region object.

[0033] In steps S4204 to S4211, the inspection conditions are set according to each object. In this embodiment, the inspection conditions include an inspection area and an inspection means. Here, the inspection area is composed of graphic information surrounding the object and coordinate information indicating the position of the object. FIG. 6 shows an example in which the inspection area of ​​each object is set. FIG. 6 may be configured to be displayed on a display unit (not shown) of the inspection PC 1500. In this embodiment, a rectangle or an ellipse is assigned as the graphic information, as in the inspection areas 6001 and 6002 in FIG. 6. Here, a rectangle or an ellipse is selected so that the area when surrounding the object is the smallest. The coordinate information is a coordinate for determining the position of the inspection area of ​​the object. When the graphic information is a rectangle, the coordinates of the top left and bottom right vertices are specified, and when the graphic information is an ellipse, the top left and bottom right figures of the rectangle that circumscribes the ellipse are specified. For example, for a rectangle 6002, coordinates (M, N) 6003 and (M+A, N+B) 6004 are held as coordinate information. Although two vertex coordinates are exemplified as the coordinate information, information for determining a unique rectangle or ellipse, such as one vertex and side length, center coordinates, and distance from the center, may be used. The inspection means is a means for optimally inspecting each object, and an inspection means corresponding to each object is set in advance. An example of an inspection method corresponding to each object in this embodiment is shown in FIG. 7. The information shown in FIG. 7 is stored in the memory unit 3203 of the inspection PC 1500, and is referred to by the control unit 3205. The information is displayed on the display unit (not shown) of the inspection PC 1500, and the settings can be changed according to a user instruction.

[0034] In step S4204, the control section S3201 determines whether the setting of inspection conditions has been completed for all objects, and if YES, the process proceeds to step S4012, and if NO, the process proceeds to step S4205.

[0035] In step S4205, the control section S3201 determines whether the target object is an image area object, and if YES, the process proceeds to step S4206, and if NO, the process proceeds to step S4209.

[0036] In step S4206, the control section S3201 determines whether the target object is an image area object, and if NO, the process proceeds to step S4207, and if NO, the process proceeds to step S4208.

[0037] In step S4207, when the target object is an image area object, the inspection condition setting for the image area object is performed. In this embodiment, as the inspection means, the structural similarity index (SSIM: Structural Similarity), which is one of the image quality evaluation indexes, is selected by referring to the image object area column 7001 in FIG. 7. Any method capable of evaluating image quality may be used, such as a difference inspection that measures a simple luminance value error between an image to be printed and a scanned image, a method that compares the position of an object by edge detection, and pattern matching, and is not limited to these forms, and various modifications and changes are possible. In this embodiment, the inspection method that measures the image quality of an image area in this way is collectively called a pattern inspection. The inspection area and inspection means of the object are notified to the CPU 4201 and stored in the storage unit 4203.

[0038] In step S4208, when the target object is an object of the element of interest, the inspection condition setting for the object of the element of interest is performed. In this embodiment, the inspection means selects the same inspection means as in step S4206 by referring to the object inspection means column 7002 of the element of interest in FIG. 7, and sets the threshold value for inspection stricter than that for the image area object. For example, the threshold value for detecting defects when performing inspection is called the inspection level, and the inspection level can be set in nine stages from 1 to 9, with inspection level 9 being the strictest inspection. In this case, if the inspection level of the image area object is 5, the inspection level of the specific object is set to 6 or more. Also, a method for evaluating image quality other than the inspection means selected in step S4206 may be selected. The inspection area and inspection means are notified to the CPU 3201 and stored in the storage unit 3203.

[0039] In step S4209, the control unit 3201 determines whether the target object is a character area object, and if YES, the process proceeds to step S4210, and if NO, the process proceeds to step S4211.

[0040] In step S4210, when the target object is a character area object, inspection conditions for the character area are set. In this embodiment, a means for evaluating character quality is selected as the inspection means. The method for evaluating character quality is selected by comparing data of the recognition result by character recognition processing (OCR: Optical Character Recognition) with data prepared in advance, with reference to the character object inspection means 7003 in FIG. 7. In addition, various modifications and changes are possible, without being limited to these forms, such as a difference inspection that measures the error in pixel values ​​between the image to be printed and the read image in the inspection area. The inspection area and inspection means are notified to the CPU 3201 and stored in the memory unit 3203.

[0041] In step S4211, the control unit 3201 sets inspection conditions for the background area object. A difference inspection is selected as the inspection method by referring to the background area object column 7004 in FIG.

[0042] In step S4006, the control unit 3001 forms an image on a sheet based on the image to be printed, and the process ends.

[0043] In step S4212, if the control unit 3201 receives a read image from the control unit 3101 of the image reading device 1300, the process proceeds to step S4213, and if not, the process proceeds to S4212.

[0044] FIG. 4B is a flowchart of the control unit 3101 of the image reading device 1300, which starts when the image reading unit 1340 reads an image.

[0045] In step S4102, if the control unit 3101 has acquired a read image, the process proceeds to step S4103, and if not, the process proceeds to step S4102.

[0046] In step S4103, the control unit 3101 transmits the acquired read image to the control unit 1160, and the process ends.

[0047] Next, in steps S4213 to S4214, inspection is performed based on the inspection conditions set in steps S4204 to S4211.

[0048] In step S4213, it is determined whether or not inspection has been completed for all inspection objects stored in storage unit 3203. If YES, the process proceeds to step S4215, and if NO, the process proceeds to step S4214.

[0049] In step S4214, inspection is performed based on the inspection condition settings stored in the storage unit 3203. Depending on the inspection means, inspection may be performed in pixel units, the entire area, or in patch units obtained by further dividing the area into smaller areas. It is determined whether the error of the read image exceeds a desired threshold value with respect to the image to be printed, and if it exceeds the desired threshold value, it is stored in the storage unit 3203 as NG, or OK if it does not exceed the desired threshold value. When inspection areas of different objects overlap, the inspection result of the object with the higher inspection priority is prioritized. In this embodiment, the inspection priority is set in the order of the object of interest, the character area object, the image area object, and the background area object, that is, the object of interest has the highest priority. For example, in an area where the object of interest and the image area object overlap, even if the inspection result of the image area object is OK, if the object of interest is NG, the inspection result of the overlapping area is determined to be NG. Also, when the inspection areas of the same object overlap, and the inspection results are different between OK and NG, the inspection result of that area is determined to be NG. In this embodiment, it is described that inspection is performed on the overlapping area and a priority is set for the output of the results, but this is not limited to this. For example, when the region of the element of interest object and the region of the image region object overlap, the overlapping region may be configured to be inspected as the region of the element of interest object according to the priority.

[0050] In step S4215, the test results stored in storage unit 3203 in step S4214 are referenced, and if YES, the process proceeds to step S4216, and if NO, the process ends as it has been determined that the necessary tests have been completed.

[0051] In step S4216, the control unit 3201 issues a NG notice, which means that an abnormality has occurred. The NG notice may be any notice that an abnormality has occurred to a user or a maintenance worker, and the form of the notice is not particularly limited. That is, the notice may be an audio notice by a human voice. The notice may be a combination of flashing or lighting of one or more lights of multiple colors of a patrol lamp. For example, the notice may be displayed on a display panel (not shown) of the image forming apparatus main body 1100. Also, the configuration may be such that the paper output destinations for printed paper that has been OK or NG are separated.

[0052] As described above, by analyzing the image to be printed, setting the inspection conditions for each object, and then performing the inspection, the user can have the inspection automatically end when inputting the print data. Furthermore, by switching the inspection method depending on the object, appropriate inspection can be performed, ensuring the reliability of the inspection accuracy.

[0053] (Another embodiment) In the previous embodiment, the input data is printed as is and inspected.

[0054] In this embodiment, a case will be described in which two images are combined by Nup processing to generate one image for printing, and then printed.

[0055] The details will be described using the flowchart in Figure 4. Here, the same components are generally given the same reference numbers and the description will be omitted. The following description will focus on the differences. In principle, the parts that are not specifically mentioned are the same as those in the above-mentioned embodiment.

[0056] When the print data for the images to be combined is received, the flow in Fig. 4A starts. In step S4002, if the print data has not been received, the flow proceeds to step S4003. In this embodiment, the print data for two sheets is received.

[0057] In step S4003, the control unit 3001 receives print setting information including at least imposition information and generates an image for printing. Imposition information indicates how multiple images are combined to be arranged on one sheet of paper, and in this embodiment, a print image 8003 is formed by combining a raster image 8001 and a raster image 8002, as shown in FIG.

[0058] As described above, when the print data input by the user is changed due to internal processing of the image forming device, if the inspection conditions are set by analyzing the raster image, the input data cannot be inspected simply. On the other hand, by using a print image that has been image-processed as in this embodiment, the finish of the print can be inspected correctly without changing the inspection process.

[0059] (Further Alternative Embodiments) In the previous embodiment, each object is acquired by analyzing the image to be printed.

[0060] In this embodiment, attribute information such as image areas and character areas is embedded in print data input by a user, and a case will be described in which the attribute information is analyzed to obtain objects.

[0061] The details will be described using the flowchart in Figure 9. Here, the same components are generally given the same reference numbers and the description will be omitted. The following description will focus on the differences. In principle, the parts that are not mentioned are the same as those in the previous embodiment.

[0062] In step S9001, the control unit 3001 analyzes attribute information for the print data before RIP processing. Attribute information refers to information such as images, characters, graphics, and lines that is included in PDL data such as PDF in order to draw objects. In this embodiment, it includes at least character and image information. The attribute information is analyzed to obtain a character area object, an image area object, and a background area object. A background area object refers to an area excluding character area objects and image area objects from the entire print data. If attribute information is assigned to only one of the images or characters, it may be possible to detect objects that could not be detected by the attribute information by performing layout analysis in step S9101.

[0063] In step S9101, a detection process of an object of an element of interest is performed on the image area object detected in step S9001. A method of designating an object of an element of interest is described with reference to FIG. 5. An object of an element of interest can be designated from an element of interest field 5001 of a print setting screen displayed on an operation unit (not shown) as in 5000. For example, the designation method may be such that an arbitrary element can be selected from options as in 5100, or such that an arbitrary element can be selected by inputting a word as in 5200. Any number and type of object of an element of interest can be selected. The method of designating an object of an element of interest is not limited to these, and for example, a type of deliverable can be selected, and when a poster or photo book is selected, a selection screen of an object of an element of interest can be displayed. On the other hand, when the image area object is not important, such as a form, the selection screen of an object of an element of interest can be switched so that it is not displayed. In addition, an image recognition model may be used to automatically determine an element of interest. When an element designated in this way is included in an image area, an inspection condition is also set for that area. A machine learning model trained to detect objects in an image is used for object detection in an image area. For example, Detectron2 (Facebook AI Research) can be mentioned.

[0064] As described above, when attribute information can be utilized, accurate inspection areas can be set by detecting objects from the attribute information. Furthermore, by detecting the object of the element of interest, more detailed inspection conditions can be set than with the technology of Patent Document 1.

Claims

1. A registration means for registering the received image as a reference image; A detection means for analyzing the image and detecting elements constituting the image; A setting means for setting inspection conditions including at least an inspection method based on the detected elements; an acquisition means for acquiring a read image by reading a printed matter on which the image is printed on a recording sheet; an inspection means for inspecting the printed matter under the inspection conditions set by the setting means, based on the read image and the reference image; An inspection device comprising:

2. If the image does not include attribute information, the detection means detects elements that constitute the image by analyzing the layout of the image.

2. The inspection apparatus according to claim 1 .

3. When the image includes attribute information, the detection means detects elements constituting the image by analyzing the attribute information.

2. The inspection apparatus according to claim 1 .

4. The inspection conditions include an inspection area, an inspection method, and an inspection level to be inspected by the inspection means.

2. The inspection apparatus according to claim 1 .

5. The inspection method is any one of a difference inspection, a pattern inspection, and an OCR inspection.

2. The inspection apparatus according to claim 1 .

6. When the detected element is a character area, the setting means sets the character area as the inspection area and sets a first inspection method as the inspection method.

5. The inspection apparatus according to claim 4.

7. When the detected element is a background area, the setting means sets the background area as the inspection area and sets a second inspection method as the inspection method.

5. The inspection apparatus according to claim 4.

8. When the detected element is an image area, the setting means sets the image area as the inspection area and sets a third inspection technique as the inspection technique.

5. The inspection apparatus according to claim 4.

9. image analysis means for further analyzing image objects depicted within the image region; A designation means for designating an attention element object to be inspected intensively; having When the detected element is an image area and the image object analyzed by the image analysis means is the specified element object of interest, the setting means sets the area of ​​the element object of interest as the inspection area and sets the inspection level to be higher than that of the image area.

5. The inspection apparatus according to claim 4.

10. When a plurality of the inspection areas set by the setting means overlap, the inspection method and the inspection level of the target element object are given priority.

10. The inspection apparatus according to claim 9.

11. The designation means designates the attention element object based on word information received from a user.

10. The inspection apparatus according to claim 9.

12. The designation means designates the target element object that is automatically determined based on the type of the product received from the user.

10. The inspection apparatus according to claim 9.

13. The detection means is a machine learning model trained to analyze the layout of an image.

3. The inspection apparatus according to claim 2.

14. A method for controlling an inspection apparatus, comprising the steps of: a registration step of registering the received image as a reference image; A setting step of analyzing the image and setting inspection conditions; an acquiring step of acquiring a read image by reading a printed matter on which the image is printed on a recording sheet; an inspection step of inspecting the printed matter under the inspection conditions set based on the read image and the reference image; A method for controlling an inspection apparatus that performs the above.

Images