Printing system and defective nozzle detection method

Abstract

To provide a printing system and a defective nozzle detecting method which can efficiently detect a defective nozzle out of a number of nozzles.SOLUTION: The printing system comprises a printing part 205 and a photographing part 310, which is further provided with: a defect detecting part 322 that detects defects included in a photographed image; a defective nozzle position candidate extracting part 323 that extracts N-position candidates as candidates for positions of defective nozzles on the basis of positions of the defects; a first correcting part 120 that performs processing for correcting the defects; and a defective nozzle identifying part 325 that identifies the defective nozzles. The first correcting part 120 performs the processing for correcting the defects, while sequentially setting only the N-nozzles corresponding to the N-position candidates as nozzles to be corrected, one by one. The defective nozzle identifying part 325 identifies the defective nozzles, on the basis of the positions of the defects detected by the defect detecting part 322 on the basis of a corrected image.SELECTED DRAWING: Figure 9

Description

Technical Field

[0001] The present invention relates to a printing system including a printing mechanism having a printing head provided with a large number of nozzles for ejecting ink and an image inspection device for inspecting an image after printing, and particularly relates to a method for detecting nozzles in a state of ejection failure from a large number of nozzles.

Background Art

[0002] Conventionally, an inkjet printing device that performs printing by ejecting ink onto a substrate (printing medium) such as printing paper is known. In an inkjet printing device, when the ejection interval becomes long, during the period of printing, drying of the ink due to evaporation of the solvent in the vicinity of the nozzle, mixing of air bubbles into the nozzle, adhesion of dust to the nozzle, etc. may occur. That is, ejection failure of the ink may occur. When ejection failure of the ink occurs, in the printed image, omission of dots corresponding to nozzles in a state of ejection failure (hereinafter referred to as "defective nozzles"), that is, dot dropout occurs. In this case, operations for restoring the function of the defective nozzles (cleaning or flushing) and alternative droplet ejection in which ink droplets to be ejected by the defective nozzles are ejected by other nozzles are performed.

[0003] In an inkjet printing device having a printing head (inkjet head) in which a large number of nozzles as recording elements are arranged in the width direction of the substrate (the direction perpendicular to the substrate conveyance direction), in order to prevent dot dropout caused by ejection failure as described above, it is necessary to detect ejection failure and identify defective nozzles. Regarding this, conventionally, defective nozzles have been identified based on the printing result of a page image consisting only of a nozzle check pattern or the printing result of an image in which a nozzle check pattern is added to an area outside the original printing area for a plurality of pages.

[0004] However, if a method for identifying defective nozzles is based on the printout of a page image consisting only of a nozzle check pattern, pages for printing the nozzle check pattern are inserted between many pages that the user needs, necessitating the removal of these pages in a later process. If a method for identifying defective nozzles is based on the printout of an image in which a nozzle check pattern is added to an area outside the original printing area for multiple pages, many pages must be processed to inspect all the nozzles in the inkjet printer, and the original printing area usable by the user is reduced by the area allocated for the nozzle check pattern.

[0005] Japanese Patent Publication No. 6945060 discloses an invention for detecting abnormal nozzles, which identifies abnormal nozzles from the print result of a user image into which an abnormal nozzle (defective nozzle) identification pattern that is not visible to the user has been embedded. According to this abnormal nozzle detection method, a correspondence is made between a partial region in the user image and a nozzle, and a correction is made assuming that the nozzle associated with each partial region is abnormal. Then, streak information is detected based on the print result of the corrected image, and the state of the nozzle is estimated based on this streak information. In this way, a defective nozzle can be identified from a single image without increasing the amount of wasted paper. Japanese Patent Application Publication No. 2005-067191 discloses a method for detecting inappropriate nozzles (defective nozzles) for K ink nozzles, C ink nozzles, M ink nozzles, and Y ink nozzles, respectively, using RGB values ​​obtained by reading a printed image. [Prior art documents] [Patent Documents]

[0006] [Patent Document 1] Patent No. 6945060 [Patent Document 2] Japanese Patent Publication No. 2005-067191 [Overview of the initiative] [Problems that the invention aims to solve]

[0007] However, according to the invention disclosed in Japanese Patent Publication No. 6945060, the entire user image is divided into multiple sub-regions, and all nozzles are associated with each sub-region one by one. Then, in all sub-regions, a correction process is performed assuming that the associated nozzle is a defective nozzle. As a result, the processing efficiency is extremely poor. Furthermore, the invention targets defect detection in sub-regions within a single user image (a user image on one page), and does not consider defect detection across multiple pages or different patterns. Moreover, no specific method for determining the sub-regions is disclosed.

[0008] In view of the above circumstances, the present invention aims to provide a printing system and a defective nozzle detection method that can efficiently detect defective nozzles from a large number of nozzles. [Means for solving the problem]

[0009] The first invention is a printing system, A printing unit having multiple nozzles, which prints on a printing medium by ejecting ink from the multiple nozzles, An imaging unit that performs imaging of the printed image printed on the printing medium by the printing unit, A defect detection unit performs inspection processing to detect defects contained in the captured image obtained by capturing the printed image by the imaging unit, A defective nozzle position candidate extraction unit extracts N (where N is an integer of 2 or more) candidate positions from the positions of the plurality of nozzles, based on the position of the defect detected by the inspection process in the captured image, to be a candidate position of a defective nozzle which is a nozzle having an ejection defect. A defect correction unit generates a corrected image by performing a defect correction process on a user image, which is the image to be printed, to remove the effects of nozzle ejection defects set for the nozzle to be corrected. A defect nozzle identification unit that identifies the defect nozzle from N nozzles corresponding to each of the N position candidates. Equipped with, After printing by the printing unit, capturing the printed image by the imaging unit, and the inspection process by the defect detection unit are performed based on the user image, the N candidate positions are extracted by the defect nozzle position candidate extraction unit. The defect correction unit sequentially sets only the N nozzles out of the plurality of nozzles as the nozzles to be corrected one by one and performs the defect correction process. After the defect correction unit generates the corrected image, the printing unit prints the corrected image, the imaging unit captures the printed image, and the defect detection unit performs the inspection process based on the corrected image. The defective nozzle identification unit identifies the defective nozzle based on the position of the defect in the captured image, which was detected by the inspection process performed based on the corrected image. death, The defect correction unit includes a mapping unit that maps the N nozzles to N partial regions included in the user image, Each of the N subregions includes the location of a defect detected by the inspection process performed based on the user image, The defect correction unit performs the defect correction process by setting different nozzles from the N nozzles to be corrected as the target nozzles for each of the N subregions. The defective nozzle identification unit identifies as the defective nozzle the nozzle corresponding to the portion of the N portion of the region in which no defect was detected by the inspection process performed based on the corrected image. It is characterized by doing so.

[0011] The 2 The invention is the first 1 In the invention, The printing unit includes a plurality of ink ejection units that eject ink of different colors from nozzles. The printing system further includes a defective color identification unit that identifies which of the plurality of ink ejection units ejects ink of which color the defective nozzle is contained in. The defect correction unit is characterized in that, when performing the defect correction process, it sets the nozzle that ejects ink of the defect color, which is the color identified by the defect color identification unit, to be the nozzle to be corrected.

[0012] The 3 The invention is the first 2 In the invention, The plurality of ink ejection units includes a black ink ejection unit that ejects black ink, a cyan ink ejection unit that ejects cyan ink, a magenta ink ejection unit that ejects magenta ink, and a yellow ink ejection unit that ejects yellow ink. The defective color specifying unit specifies the defective color from colors other than the color of ink for which the amount to be ejected at the position of a defect detected by the inspection process performed based on the user image among black ink, cyan ink, magenta ink, and yellow ink is less than or equal to a predetermined threshold value.

[0013] No. 4 The invention according to 2 or 3 In the invention according to the defective color specifying unit specifies the defective color based on the difference in the average value of the red color values, the difference in the average value of the green color values, and the difference in the average value of the blue color values in a region where a defect has been detected by the inspection process performed based on the user image, between the captured image based on the user image and the correct image corresponding to the user image.

[0014] No. 5 The invention according to 2 to 4 In any of the inventions according to the association unit sets, as a processing target region, a region among the regions corresponding to the N position candidates in which the density value of the defective color satisfies a predetermined condition, and associates the N nozzles with the N partial regions in the processing target region.

[0015] No. 6 The invention according to 5 In the invention according to the association unit when the size of each partial region is greater than or equal to the size of the region required for defect detection by the inspection process when the processing target region within one page is divided into N, associates the N nozzles with the N partial regions obtained by dividing the processing target region within one page into N. When the processing area within a page is divided into N parts, if the size of each sub-area is less than the size of the area required for defect detection by the inspection process, the processing area within a page is divided into M sub-areas (where M is an integer less than N) within a range where the size of each sub-area after division is equal to or greater than the size of the area required for defect detection by the inspection process, and M nozzles out of the N nozzles are associated with the M sub-areas.

[0016] The 7 The invention is the first 1 From the first 6 In any of the inventions up to this point, The aforementioned user image is an image consisting of two or more pages, The correspondence unit is characterized by distributing the N sub-regions across multiple pages and associating the N nozzles with the N sub-regions.

[0017] The 8 The invention is the first 7 In the invention, The size of each of the N subregions is characterized by being greater than or equal to the size of the region required for defect detection by the inspection process.

[0018] The ninth invention is, in any of the first to eighth inventions, The defective nozzle position candidate extraction unit is characterized by converting the coordinates of the position of the defect detected by the inspection process performed based on the user image into coordinates on the user image, and extracting the N position candidates from the positions of the plurality of nozzles based on the converted coordinates. 。

[0019] The 10 The invention relates to a method for detecting a defective nozzle in a printing system, which includes a printing unit having a plurality of nozzles and performing printing on a printing medium by ejecting ink from the plurality of nozzles, and an imaging unit that captures a printed image printed on the printing medium by the printing unit, The first printing step involves the printing unit printing a user image, A first imaging step in which the printed image obtained in the first printing step is captured, A first defect detection step for detecting defects included in the first image obtained in the first imaging step, A defective nozzle position candidate extraction step is performed to extract N (where N is an integer of 2 or more) candidate positions from the positions of a plurality of nozzles as candidate positions for a defective nozzle having an ejection defect, based on the position of the defect detected in the first defect detection step in the first image capture; A defect correction step to generate a corrected image by performing a defect correction process on the user image to remove the effects of nozzle discharge defects set for the nozzle to be corrected, A second printing step in which the printing unit prints the corrected image, A second imaging step for capturing the printed image obtained in the second printing step, A second defect detection step for detecting defects included in the second image obtained in the second imaging step, A defect nozzle identification step involves identifying the defect nozzle from N nozzles corresponding to each of the N position candidates based on the position of the defect detected in the second defect detection step in the second captured image, and Includes, In the defect correction step, only the N nozzles out of the plurality of nozzles are sequentially set one by one as the nozzles to be corrected, and the defect correction process is performed. The defect correction step includes a mapping step that associates the N nozzles with N subregions included in the user image, Each of the N subregions includes the location of the defect detected in the first defect detection step, In the defect correction step, for each of the N subregions, different nozzles from among the N nozzles are set as the nozzles to be corrected, and the defect correction process is performed. The defective nozzle identification step is characterized in that the nozzles associated with the subregions among the N subregions in which no defects were detected in the second defect detection step are identified as the defective nozzles. 。

[0020] Also, the 10 Modifications that can be understood by referring to the embodiments and drawings of the invention are considered to be means for solving the problem. [Effects of the Invention]

[0021] According to the first invention described above, based on the location of defects detected in an inspection process based on an image obtained by capturing a printed image of a user image, N (where N is an integer of 2 or more) candidate positions for the defective nozzle are extracted. Then, only the N nozzles corresponding to each of these N candidate positions are sequentially set as nozzles to be corrected, one by one, and a defect correction process (a process to remove the effect of ejection defects in the nozzles set as nozzles to be corrected) is performed. After that, the defective nozzle is identified based on the location of defects detected in an inspection process based on an image obtained by capturing a corrected image (an image after the defect correction process). As described above, only N (for example, 4) nozzles corresponding to each of the N candidate positions are set as nozzles to be corrected during the defect correction process from among the multiple nozzles provided in the printing system. Furthermore, defective nozzles can be identified based on the results of only N subregions from the inspection process performed based on the corrected image. Furthermore, it is possible to identify defective nozzles without printing special images such as nozzle check patterns. extremely This will enable a printing system that can efficiently detect defective nozzles.

[0023] The above 2 According to this invention, a color printing system capable of efficiently detecting defective nozzles from a large number of nozzles is realized.

[0024] The above 3 According to this invention, it is possible to effectively prevent the misidentification of a non-defective color as a defective color.

[0025] The above 4 According to this invention, it is possible to accurately identify which color ink nozzle has a defect.

[0026] The above 5 According to this invention, the difference between the region where an actual defective nozzle is set as the nozzle to be corrected and the region where a nozzle that is not actually a defective nozzle is set as the nozzle to be corrected becomes clear in the captured image obtained by capturing an image after defect correction processing (corrected image), so that the actual defective nozzle can be identified with high accuracy.

[0027] The above 6 According to this invention, defective nozzles can be accurately identified regardless of how defects (streaks) appear in the printed image.

[0028] The above 7 According to this invention, even if only short defects (streak defects) occur within a single page of printed image, it is possible to identify the defective nozzle by processing multiple pages.

[0029] The above 8 According to the invention, the above-mentioned 7 The same effect as the previous invention can be obtained.

[0030] The above 9 According to this invention, even if there is a misalignment in the printing position on the printing medium, candidate locations for defective nozzles can be extracted with high accuracy.

[0031] According to the 11th invention described above, the same effects as those of the first invention described above can be obtained. [Brief explanation of the drawing]

[0032] [Figure 1] This is an overall configuration diagram of a printing system according to one embodiment of the present invention. [Figure 2] This is a schematic diagram showing one example of the configuration of the inkjet printing apparatus in the above embodiment. [Figure 3] This is a plan view showing one example of the configuration of the printing section in the above embodiment. [Figure 4] This is a plan view showing one example configuration of a print head in the above embodiment. [Figure 5] This block diagram shows the configuration of the computer included in the printing system according to the above embodiment. [Figure 6] This figure illustrates the schematic of defect nozzle identification in the above embodiment. [Figure 7] This figure illustrates the schematic of defect nozzle identification in the above embodiment. [Figure 8] This figure illustrates the schematic of defect nozzle identification in the above embodiment. [Figure 9] This is a functional block diagram showing the functional configuration related to the defective nozzle detection process in the above embodiment. [Figure 10] In the above embodiment, the flowchart shows the general procedure for the defective nozzle detection process. [Figure 11] In the above embodiment, the flowchart shows the detailed procedure for the defect nozzle location candidate extraction process. [Figure 12] This is a diagram illustrating the comparison by shaking in the above embodiment. [Figure 13] This figure illustrates the defect list in the above embodiment. [Figure 14] This figure illustrates the defect list in the above embodiment. [Figure 15] In the above embodiment, the flowchart shows the detailed procedure for identifying defective colors. [Figure 16] In the above embodiment, the flowchart shows the detailed procedure for the target color narrowing process. [Figure 17] This figure illustrates the processing performed for each unit region in the above embodiment. [Figure 18]In the above embodiment, the flowchart shows the detailed procedure for the defect analysis process. [Figure 19] This figure illustrates the defect color determination table in the above embodiment. [Figure 20] In the above embodiment, the flowchart shows the detailed procedure of the correction process performed by the first correction unit. [Figure 21] This figure illustrates the correspondence between a partial region and a nozzle in the above embodiment. [Figure 22] This figure illustrates the correspondence between a partial region and a nozzle in the above embodiment. [Figure 23] This figure illustrates the correspondence between a partial region and a nozzle in the above embodiment. [Figure 24] In the above embodiment, the flowchart shows the detailed procedure for identifying defective nozzles. [Modes for carrying out the invention]

[0033] An embodiment of the present invention will be described below with reference to the attached drawings.

[0034] <1. Overall configuration of the printing system> Figure 1 is an overall configuration diagram of a printing system according to one embodiment of the present invention. This printing system consists of an inkjet printing device 10 and a print data generation device 40. The inkjet printing device 10 and the print data generation device 40 are connected to each other by a communication line 5. The print data generation device 40 generates print data by performing RIP processing on input data such as PDF files. The print data generated by the print data generation device 40 is transmitted to the inkjet printing device 10 via the communication line 5. The inkjet printing device 10 outputs a print image onto printing paper, which is the printing medium, based on the print data transmitted from the print data generation device 40, without using a printing plate. The inkjet printing device 10 consists of a printing press body 200, a printing control device 100 that controls the operation of the printing press body 200, and an image inspection device 300 that inspects the printing status. In other words, this inkjet printing device 10 is a printing device with an inspection function. Some components of the image inspection device 300 are built into the printing press body 200.

[0035] In the configuration shown in Figure 1, the image inspection device 300 is a component of the inkjet printing device 10 (i.e., the image inspection device 300 is included in the inkjet printing device 10), but this is not the only option. The image inspection device 300 may be a separate device independent of the inkjet printing device 10.

[0036] <2. Configuration of an inkjet printing device> Figure 2 is a schematic diagram showing one example configuration of an inkjet printing apparatus 10. As described above, this inkjet printing apparatus 10 is composed of a printing control device 100, a printing machine body 200, and an image inspection device 300. The printing machine body 200 includes a paper delivery unit 202 for supplying printing paper (e.g., roll paper) PA, a printing mechanism 201 for printing on the printing paper PA, and a paper winding unit 208 for winding up the printed printing paper PA. The printing mechanism 201 includes a first drive roller 203 for transporting the printing paper PA into the mechanism, a plurality of support rollers 204 for transporting the printing paper PA inside the printing mechanism 201, a printing unit 205 for ejecting ink onto the printing paper PA to perform printing, a drying unit 206 for drying the printed printing paper PA, and a second drive roller 207 for outputting the printing paper PA from inside the printing mechanism 201.

[0037] The first drive roller 203, the multiple support rollers 204, and the second drive roller 207 constitute a transport mechanism for moving the printing paper PA. The printing unit 205 includes print heads 25K, 25C, 25M, and 25Y, which eject black (K), cyan (C), magenta (M), and yellow (Y) inks, respectively. Inside the printing mechanism 201 is an imaging unit 310 that captures the printed image formed on the printing paper PA by the printing unit 205. The imaging unit 310 is a component of the image inspection device 300 and is configured using an image sensor such as a CCD or CMOS.

[0038] The print control device 100 controls the operation of the printing press body 200, which has the configuration described above. When the print control device 100 is given a command to print output, the print control device 100 controls the operation of the printing press body 200 so that the printing paper PA is transported from the paper feeding unit 202 to the paper winding unit 208. First, the printing unit 205 prints on the printing paper PA based on the printing data transmitted from the print data generation device 40, then the drying unit 206 dries the printing paper PA, and finally the imaging unit 310 captures the printed image.

[0039] The image inspection device 300 consists of an imaging unit 310 and an image inspection computer 320. The image obtained by imaging the printed image with the imaging unit 310 is sent to the image inspection computer 320. The image inspection computer 320 performs a series of processes to detect defects and identify defective nozzles, as will be described later. During this process, data necessary for processing is exchanged between the print control device 100 and the image inspection device 300.

[0040] Figure 3 is a plan view showing one example configuration of the printing unit 205. As shown in Figure 3, the printing unit 205 consists of black, cyan, magenta, and yellow print heads (ink ejection units) 25K, 25C, 25M, and 25Y arranged in a row in the transport direction (sub-scanning direction) of the printing paper PA. Each print head 25 is composed of multiple head modules 251 arranged in a staggered pattern.

[0041] Figure 4 is a plan view showing one example configuration of a single print head 25. As shown in Figure 4, the print head 25 is composed of multiple head modules 251. Each head module 251 is equipped with multiple nozzles 252 in at least the main scanning direction. Each head module 251 also has a built-in head memory 253. The head memory 253 stores unique information of the head module 251.

[0042] In the example shown in Figure 4, the print head 25 is composed of five head modules 251(1) to 251(5). Head module 251(1) is located downstream in the transport direction of the printing paper PA and on the leftmost side in the main scanning direction. Head module 251(5) is located downstream in the transport direction of the printing paper PA and on the rightmost side in the main scanning direction. Head module 251(3) is located between head module 251(1) and head module 251(5). Upstream in the transport direction of the printing paper PA, head modules 251(2) and 251(4) are arranged in a staggered pattern with the five head modules 251(1) to 251(5).

[0043] <3. Computer Hardware Configuration> Figure 5 is a block diagram showing the configuration of the computer 500 included in the printing system according to this embodiment. The computer 500 is included in the print control device 100, the image inspection device 300, and the print data generation device 40. The computer 500 included in the image inspection device 300 is the image inspection computer 320 described above.

[0044] The computer 500 shown in Figure 5 comprises a main unit 510, an auxiliary storage device 521, an optical disc drive 522, a display unit 523, a keyboard 524, and a mouse 525. The main unit 510 includes a CPU 511, memory 512, a first disk interface unit 513, a second disk interface unit 514, a display control unit 515, an input interface unit 516, and a network interface unit 517. The CPU 511, memory 512, first disk interface unit 513, second disk interface unit 514, display control unit 515, input interface unit 516, and network interface unit 517 are connected to each other via a system bus. The auxiliary storage device 521 is connected to the first disk interface unit 513. The auxiliary storage device 521 is a magnetic disk drive or the like. The optical disc drive 522 is connected to the second disk interface unit 514. An optical disc 59, such as a CD-ROM or DVD-ROM, which is a computer-readable recording medium, is inserted into the optical disc drive 522. A display unit (display device) 523 is connected to the display control unit 515. The display unit 523 is a liquid crystal display or the like. The display unit 523 is used to display information desired by the operator. A keyboard 524 and a mouse 525 are connected to the input interface unit 516. The keyboard 524 and mouse 525 are used by the operator to input instructions to this computer 500. The network interface unit 517 is a wired or wireless communication interface circuit and is connected to the communication line 5.

[0045] The auxiliary storage device 521 stores programs to be executed by the computer 500. The CPU 511 implements various functions by reading the programs stored in the auxiliary storage device 521 into the memory 512 and executing them. The memory 512 includes RAM and ROM. The memory 512 functions as a work area for the CPU 511 to execute the programs stored in the auxiliary storage device 521. The programs are provided, for example, stored in the above-mentioned computer-readable recording medium (non-transient recording medium).

[0046] <4. Outline of Defective Nozzle Identification> Referring to Figures 6 to 8, the method for identifying defective nozzles (nozzles in a discharge failure state) in this embodiment will be outlined. Here, we will take as an example the case in which a streak defect, denoted by reference numeral 60 in Figure 6, is detected by the inspection process of the image inspection device 300, which detects defects (flaws in the printed image). Hereafter, the process of setting a nozzle that is suspected to be causing the streak defect as a nozzle to be corrected and correcting the image based on the print data so that a printed image is obtained in which the discharge defect of the nozzle to be corrected has been removed will be referred to as the "defect correction process".

[0047] First, based on the location of the streak defect 60 (coordinates in the main inspection direction), candidate nozzles causing the streak defect 60 (hereinafter referred to as "candidate defective nozzles") are extracted. Since the resolution of the captured image is lower than the resolution of the printed image, for example, four nozzles are extracted as candidate defective nozzles. Incidentally, one nozzle and another are identified by their arrangement positions. Therefore, for example, four positions are extracted as candidate positions (candidate arrangement positions for the nozzles causing the streak defect 60). Once four positions are extracted as candidate positions, the area where the streak defect 60 occurs is divided into four sub-regions 61 to 64, as shown in Figure 7. Here, for the sake of explanation, the four nozzles extracted as candidate defective nozzles are referred to as the "first nozzle," the "second nozzle," the "third nozzle," and the "fourth nozzle." For partial region 61, the first nozzle is set as the nozzle to be corrected and defect correction processing is performed; for partial region 62, the second nozzle is set as the nozzle to be corrected and defect correction processing is performed; for partial region 63, the third nozzle is set as the nozzle to be corrected and defect correction processing is performed; and for partial region 64, the fourth nozzle is set as the nozzle to be corrected and defect correction processing is performed. Let's assume that the printing result shown in Figure 8 is obtained after the defect correction processing. According to the printing result shown in Figure 8, streak defects remain in partial regions 61, 62, and 64, but the streak defects are eliminated in partial region 63. In this case, in the inspection process after the defect correction processing, streak defects are detected in partial regions 61, 62, and 64, but not in partial region 63. In relation to this, even if the defect correction processing is performed with a nozzle that does not cause streak defects set as the nozzle to be corrected, the streak defects are not eliminated, but if the defect correction processing is performed with a nozzle that causes streak defects set as the nozzle to be corrected, the streak defects are eliminated. Therefore, in the example shown in Figure 8, the third nozzle is identified as the defective nozzle that is actually causing the streak defect, based on the results of the inspection process after the defect correction process.

[0048] In addition, the inspection process by the image inspection device 300 also detects defects caused by events other than nozzle ejection failure (for example, ink dripping), but in this embodiment, we will focus only on streak defects. Furthermore, in the following, the process mainly performed by the image inspection computer 320 and the print control device 100 to identify a defective nozzle from among the multiple nozzles provided in the printing unit 205 will be referred to as the "defective nozzle detection process."

[0049] <5. Functional Configuration> Figure 9 is a functional block diagram showing the functional configuration related to the defective nozzle detection process in this embodiment. The print control device 100 is provided with a source image storage unit 110, a first correction unit 120, and a second correction unit 130 as functional components related to the defective nozzle detection process. The image inspection computer 320 is provided with a correct answer data creation unit 321, a defect detection unit 322, a defective nozzle position candidate extraction unit 323, a defect color identification unit 324, and a defective nozzle identification unit 325 as functional components related to the defective nozzle detection process. The operation of each component will be described below in accordance with the flow of the defective nozzle detection process.

[0050] The original image storage unit 110 stores the user image, which is the image to be printed. This user image is the image before shading is applied and corresponds to the print data transmitted from the print data generation device 40. Here, "user image" refers to an image specified by the user in order to print the image the user desires, and is synonymous with "actual image," such as the product image that the user wants to print, or a test image used for test printing of the product image in advance.

[0051] The correct answer data creation unit 321 receives the user image stored in the original image storage unit 110 from the print control device 100. The correct answer data creation unit 321 then creates a correct answer image in RGB format from the user image in CMYK format.

[0052] The defect detection unit 322 performs inspection processing to detect defects contained in the captured image obtained by the imaging unit 310 capturing the printed image. This inspection processing is performed by comparing the captured image with the correct image created by the correct data creation unit 321. In this regard, since the captured image is in RGB format, the correct data creation unit 321 creates a correct image in RGB format as described above. The inspection processing by the defect detection unit 322 is also performed by comparing the captured image of the printed image based on the user image with the correct image, and also by comparing the captured image of the printed image based on the correction image described later with the correct image.

[0053] The defective nozzle position candidate extraction unit 323, while referring to the captured image and the ground truth image, extracts candidate positions for defective nozzles (position candidates) from the positions of multiple nozzles provided in the printing unit 205, based on the position of the defect detected by the inspection process by the defect detection unit 322 (position in the captured image: coordinates in the main scanning direction). In this embodiment, with N being an integer of 2 or more, the defective nozzle position candidate extraction unit 323 extracts N position candidates. In other words, the defective nozzle position candidate extraction unit 323 extracts N nozzles as defective nozzle candidates from among the multiple nozzles provided in the printing unit 205.

[0054] The defective color identification unit 324 identifies which of the following print heads—black print head 25K, cyan print head 25C, magenta print head 25M, and yellow print head 25Y—is ejecting a defective nozzle. In general terms, the defective color identification unit 324 identifies the color based on the difference in the average values ​​of R (red), G (green), and B (blue) color values ​​in the region where defects were detected by the inspection process based on the user image, between the captured image based on the user image and the ground truth image corresponding to the user image. At this time, the defective color identification unit 324 narrows down the color based on data from the portions of the user image corresponding to the N position candidates, and further identifies the color based on data from the captured image and the ground truth image corresponding to the N position candidates. Hereinafter, the color identified by this defective color identification unit 324 will be referred to as the "defective color."

[0055] The first correction unit 120 receives information on candidate positions extracted by the defective nozzle position candidate extraction unit 323 and information on the defective color identified by the defective color identification unit 324 from the image inspection computer 320. Based on the information on candidate positions and the defective color, the first correction unit 120 sequentially sets only the N nozzles (nozzles extracted as defective nozzle candidates by the defective nozzle position candidate extraction unit 323) from among the multiple nozzles provided in the printing unit 205 as nozzles to be corrected one by one, and generates a corrected image by performing the above-described defect correction process on the user image. After a defective nozzle corresponding to a certain streak defect has been identified, the second correction unit 130 performs a defect correction process (defect correction process to remove the effect of the ejection defect of the defective nozzle) on the user image, and the image obtained (corrected user image) is provided to the first correction unit 120, and the first correction unit 120 generates a corrected image by performing a defect correction process on the corrected user image.

[0056] The first correction unit 120 includes a mapping unit 122. The mapping unit 122 maps N nozzles, which are candidates for defective nozzles, to N sub-regions included in the user image. At that time, the N sub-regions are set up so that the location of the defect detected by the inspection process, which is performed by comparing the captured image of the print image based on the user image with the correct image, is included in each sub-region. The defect correction process by the first correction unit 120 is performed by setting a different nozzle from the N nozzles to be corrected for each of these N sub-regions.

[0057] The defective nozzle identification unit 325 is provided with correction position information from the print control device 100, indicating which nozzle was set as the target nozzle for defect correction processing at which location (region). The defective nozzle identification unit 325 then identifies the defective nozzle based on the correction position information and the location of the defect (location in the captured image: coordinates in the main scanning direction) detected in the inspection process, which is performed by comparing the captured image of the print image based on the correction image with the correct image.

[0058] The second correction unit 130 receives information about the defective nozzles identified by the defective nozzle identification unit 325 from the image inspection computer 320. The second correction unit 130 then performs defect correction processing on the user image so that a printed image is obtained in which the effects of the ejection defects of the defective nozzles are removed. This results in the corrected user image described above. After all defective nozzles have been identified, printing is performed by the printing unit 205 based on the corrected user image generated by the second correction unit 130.

[0059] Incidentally, as mentioned above, the user image stored in the original image storage unit 110 is the image before shading is applied. Therefore, the print control device 100 is also provided with components for performing shading so that the print unit 205 is provided with image data after shading. However, these components are omitted in Figure 9.

[0060] <6. Processing Procedure> The procedure for detecting defective nozzles is described below.

[0061] <6.1 Outline Procedure> Figure 10 is a flowchart outlining the procedure for detecting defective nozzles. Since this defective nozzle detection process is repeated, the procedure described here represents one page of the process.

[0062] First, the print control device 100 receives a user image transmitted from the print data generation device 40 (step S10). The user image is stored in the original image storage unit 110. Next, the user image stored in the original image storage unit 110 is transmitted from the print control device 100 to the image inspection computer 320, and the correct answer data creation unit 321 creates a correct answer image from the user image (step S11).

[0063] Next, the printing unit 205 prints based on the user image stored in the original image storage unit 110 (step S12). Then, the printed image obtained in step S12 is captured by the imaging unit 310 (step S13). Next, the defect detection unit 322 performs an inspection process to detect defects in the captured image by comparing the captured image obtained in step S13 with the correct image created in step S11 (step S14).

[0064] Next, the defective nozzle position candidate extraction unit 323 performs a defective nozzle position candidate extraction process to extract the above-mentioned N position candidates (candidate positions for defective nozzles) from the positions of multiple nozzles provided in the printing unit 205 (step S15). Details of the defective nozzle position candidate extraction process will be described later. Next, the defective color identification unit 324 performs a defective color identification process to identify the defective color (step S16). Details of the defective color identification process will be described later.

[0065] Next, the first correction unit 120 performs a correction process (step S17). This correction process includes defect correction, and as described above, only N nozzles out of the multiple nozzles provided in the printing unit 205 that are candidates for defective nozzles are sequentially set as nozzles to be corrected one by one, and the defect correction process is performed. This generates a corrected image. Details of the correction process performed in step S17 will be described later.

[0066] Next, printing is performed by the printing unit 205 based on the corrected image generated in step S17 (step S18). Then, the printed image obtained in step S18 is captured by the imaging unit 310 (step S19). Next, the defect detection unit 322 performs an inspection process to detect defects in the captured image by comparing the captured image obtained in step S19 with the correct image created in step S11 (step S20).

[0067] Next, the defective nozzle identification unit 325 performs a defective nozzle identification process (step S21) to identify the defective nozzle based on the location of the defect (location in the captured image) detected in step S20. Details of the defective nozzle identification process will be described later.

[0068] Next, correction processing is performed by the second correction unit 130 (step S22). Specifically, defect correction processing is performed on the user image so that a printed image is obtained in which the effects of the ejection defects of the defective nozzles identified in step S21 have been removed. This results in the corrected user image described above. If, after a defective nozzle corresponding to a certain streak defect has been identified in step S21, defect correction processing is performed by the first correction unit 120 to identify a defective nozzle causing another streak defect, the corrected user image generated in step S22 is provided to the first correction unit 120. After all defective nozzles have been identified, printing is performed by the printing unit 205 based on the corrected user image generated in step S22.

[0069] In this embodiment, the first printing step is realized by step S12, the first imaging step is realized by step S13, the first defect detection step is realized by step S14, the defect nozzle position candidate extraction step is realized by step S15, the defect color identification step is realized by step S16, the defect correction step is realized by step S17, the second printing step is realized by step S18, the second imaging step is realized by step S19, the second defect detection step is realized by step S20, and the defect nozzle identification step is realized by step S21.

[0070] <6.2 Defect nozzle location candidate extraction process> Figure 11 is a flowchart detailing the procedure for extracting candidate defective nozzle locations. After the defective nozzle location extraction process begins, the ground truth image (original image) is first divided into multiple blocks of a predetermined range, and a process called "shaking comparison" is performed for each block, in which the relative positional relationship between the ground truth image and the captured image is slightly shifted and compared (step S150). This process is performed to align the ground truth image and the captured image because there is a shift in the position where printing is performed on the printing paper. In step S150, for example, as shown in Figure 12, nine cases (cases 1 to 9) are prepared regarding the relative positional relationship between the ground truth image 71 and the captured image 72, and alignment is performed corresponding to the case in which the difference in color values ​​(RGB values) between the two is minimized.

[0071] Next, the location (coordinates) of the defect in the captured image is converted to a location (coordinates) on the user image (step S151). Then, based on the converted coordinates (coordinates in the main scanning direction), N candidate locations for the defective nozzle are determined (step S152). In other words, the location of the candidate defective nozzle is identified. In this regard, for example, information relating the coordinates in the main scanning direction on the user image to the location (coordinates in the main scanning direction) of each nozzle is stored in advance, and based on this information, the location (coordinates) of the candidate defective nozzle is identified from the coordinates obtained in step S151.

[0072] After the location of a candidate defective nozzle is identified, it is determined whether the location of the candidate defective nozzle (coordinates in the main scanning direction) is already stored in a list that holds information on streak defects (hereinafter referred to as the "defect list") (step S153). If the coordinates in the main scanning direction are stored in the defect list, the process proceeds to step S154; otherwise, the process proceeds to step S155.

[0073] In step S154, the information regarding the length of the relevant streak defect in the defect list is corrected. In step S155, the information regarding the relevant streak defect is added to the defect list. With respect to steps S154 and S155, the processing in step S152 is performed block by block. For example, if a streak defect occurs across two blocks, the information regarding the streak defect is added to the defect list during the processing of the first block, and the information regarding the length of the streak defect in the defect list is corrected during the processing of the second block. After the completion of step S154 or step S155, this defect nozzle position candidate extraction process ends. For example, if two streak defects occur on the page being processed as shown in Figure 13, the information regarding both streak defects will be retained in the defect list as shown in Figure 14 when this defect nozzle position candidate extraction process ends.

[0074] <6.3 Defect Color Identification Process> Figure 15 is a flowchart detailing the procedure for identifying defective colors. After the start of the defective color identification process, it is first determined whether or not defect information (information on the location of candidate defective nozzles and information on the defective color) has already been sent to the print control device 100 (step S161). If the defect information has been sent to the print control device 100, the defective color identification process ends; if the defect information has not been sent to the print control device 100, the process proceeds to the first loop. In this regard, for example, if the same streak defect occurs on both the first and second pages due to a single defective nozzle, the defect information will be sent to the print control device 100 during the processing of the first page, but the defective color identification process will end during the processing of the second page without the defect information being sent to the print control device 100.

[0075] Regarding the processing of the first loop, in step S162, a target color narrowing process (a process to narrow down the colors targeted for processing in the second loop) is performed to narrow down the candidate colors for the defective color from black, cyan, magenta, and yellow, based on the user image and information on the location of the candidate defective nozzle. Details of the target color narrowing process will be described later.

[0076] Subsequently, a determination is made (step S163) as to whether three colors have been excluded from the list of candidate defective colors through the target color narrowing process. If three colors have been excluded from the list of candidate defective colors, the process proceeds to step S166; otherwise, the process proceeds to the second loop. If three colors have been excluded from the list of candidate defective colors, the remaining color among black, cyan, magenta, and yellow is designated as the defective color.

[0077] Regarding the second loop, in step S164, a defect analysis process is performed to analyze defects in detail based on the ground truth image and the captured image. Details of the defect analysis process will be described later. After the completion of the second loop process, the defect color is determined based on the results of the defect analysis process (step S165). In this regard, as will be described later, in the second loop process, the defect color is determined for each unit area in the region corresponding to the streak defect. Then, in step S165, the color that was most frequently determined as the defect color in the second loop process is determined as the final defect color.

[0078] In step S166, based on the results of the defective nozzle position candidate extraction process described above and the results of step S165, defect information is transmitted from the image inspection computer 320 to the print control device 100. Subsequently, the defect information is stored internally in the image inspection computer 320 (for example, in the auxiliary storage device 521) (step S167), and this defect color identification process ends. When the defect correction process is performed by the first correction unit 120, the nozzle that ejects ink of the color (defect color) identified in this defect color identification process is set as the nozzle to be corrected.

[0079] Figure 16 is a flowchart showing the detailed steps of the target color filtering process. Note that this target color filtering process is part of the first loop, as shown in Figure 15, and is therefore repeated multiple times.

[0080] After the target color filtering process begins, first, based on the user image and information on the location of the candidate defective nozzles, the maximum value among the density values ​​of multiple pixels contained in the unit area described below is obtained for each of the colors black, cyan, magenta, and yellow (step S1621).

[0081] Here, we will explain the unit region with reference to Figure 17. In this embodiment, a region composed of a predetermined number of pixels that are consecutive in the sub-scanning direction is treated as a unit region. For example, a region composed of three pixels that are consecutive in the sub-scanning direction is treated as a unit region. In this case, as shown in Figure 17, if a streak defect containing n pixels P(1) to P(n) occurs, pixels P(1) to P(3) are treated as the target unit region during the first processing. Then, pixels P(2) to P(4) are treated as the target unit region during the second processing, and pixels P(3) to P(5) are treated as the target unit region during the third processing. Subsequently, pixels P(n-2) to P(n) are treated as the target unit region during the (n-2)th processing. In this way, the processing of the first loop is carried out while gradually shifting the range of pixels treated as a unit region. Note that in the above example, for example, pixels P(3) to P(5) may be treated as the target unit region during the second processing.

[0082] Therefore, in step S1621, for each of black, cyan, magenta, and yellow, the maximum value among, for example, three pixel density values ​​is obtained. Thus, in step S1621, four maximum values ​​are obtained (the maximum value for black, the maximum value for cyan, the maximum value for magenta, and the maximum value for yellow).

[0083] Next, each of the four maximum values ​​obtained in step S1621 is compared with a predetermined threshold, and colors whose maximum value is less than or equal to the threshold are excluded from the target colors (candidate defective colors) (step S1622). In other words, colors with very small density values ​​are excluded from the target colors (candidate defective colors).

[0084] Therefore, among the black, cyan, magenta, and yellow inks, the defective color will be identified from colors other than those whose amount to be ejected at the location of the defect detected by the inspection process based on the user image is below a predetermined threshold.

[0085] By the way, as mentioned above, the process consisting of steps S1621 and S1622 is repeated multiple times. In the example shown in Figure 17, if n is 50, the process consisting of steps S1621 and S1622 is repeated 48 times. In this embodiment, any color whose maximum value falls below the threshold at least once during the process of step S1622, which has been repeated multiple times, is excluded from the target colors, but this is not limited to this. For example, any color whose maximum value falls below the threshold at a predetermined rate or higher during the process of step S1622, which has been repeated multiple times, may be excluded from the target colors.

[0086] Figure 18 is a flowchart showing the detailed steps of the defect analysis process. Note that this defect analysis process is part of the second loop, as shown in Figure 15, and is therefore repeated multiple times. In the example shown in Figure 17, if n is 50, this defect analysis process will be repeated 48 times.

[0087] After the defect analysis process begins, in step S13 of the image obtained in Figure 10, the average values ​​of the color values ​​of multiple pixels contained in a unit area are calculated for each of R (red), G (green), and B (blue) in the region where streaky defects occur (step S1641). In other words, in step S1641, the average values ​​for R, G, and B are calculated based on the image.

[0088] Next, for the region in the ground truth image created in step S11 where streaky defects occur, the average value of the color values ​​of multiple pixels contained in the unit region is calculated for each of R, G, and B (step S1642). That is, in step S1642, the average value for R, the average value for G, and the average value for B are calculated based on the ground truth image.

[0089] Next, for each of R, G, and B, the difference between the mean value calculated in step S1641 and the mean value calculated in step S1642 is calculated as the amount of variation (step S1643).

[0090] Next, based on the variation amount of each color obtained in step S1643 and a pre-prepared defective color determination table, defective colors are determined for the unit area being processed (step S1644). At this time, colors whose variation amount calculated in step S1643 is greater than or equal to a predetermined threshold are treated as "Variation Amount: Large". The following explains how defective colors are determined based on the defective color determination table.

[0091] Figure 19 is a schematic diagram showing an example of a defective color determination table. The defective color determination table holds information for identifying a color to be determined as a defective color in step S1644 from among the process colors (black, cyan, magenta, and yellow) in accordance with the results obtained in step S1643 for R, G, and B, for each of two or more combinations of colors from black, cyan, magenta, and yellow. The section labeled 73 stores possible combinations of colors that were not excluded from the target colors in the processing of the first loop. The section labeled 74 stores information corresponding to the variation amount for each of R, G, and B. The section labeled 75 stores information on the color to be determined as a defective color. Furthermore, a circle in the section marked with code 74 indicates that the corresponding color has a "large variation," a blank space in the section marked with code 74 indicates that the corresponding color does not have a "large variation," a hyphen in the section marked with code 74 indicates that there are no cases in which the corresponding color has a "large variation," and a hyphen in the section marked with code 75 indicates that the corresponding case will not occur. Additionally, the "X" in the section marked with code 75 indicates that, among multiple colors with a "large variation," only the color with the largest variation should be considered to have a "large variation" when determining a defective color.

[0092] For example, suppose a defective color determination table like the one shown in Figure 19 is available, and that none of the colors were excluded from the target colors (candidate defective colors) during the processing of the first loop. In this case, the target colors are cyan (C), magenta (M), yellow (Y), and black (K). If R, G, and B all have a "large variation," then black (K) is determined to be a defective color by referring to the part indicated by the arrow labeled with symbol 76.

[0093] Furthermore, suppose, for example, that a defect color determination table like the one shown in Figure 19 is provided, and that in the first loop processing, black (K) is excluded from the list of target colors (candidates for defect colors). In this case, the target colors are cyan (C), magenta (M), and yellow (Y). If only G has a "large variation," then magenta (M) is determined to be a defect color by referring to the part indicated by the arrow labeled with symbol 77.

[0094] Furthermore, let's assume, for example, that a defective color determination table like the one shown in Figure 19 is provided, and that none of the colors were excluded from the list of candidate target colors (candidate defective colors) during the processing of the first loop. In this case, the target colors are cyan (C), magenta (M), yellow (Y), and black (K). If G and B have a "large variation amount," and the variation amount of G is greater than that of B, then magenta (M) is determined to be a defective color by referring to the parts indicated by the arrows labeled 78 and 79.

[0095] As described above, in step S1644, the defective color is determined using the defective color determination table. This defect analysis process is repeated multiple times. Therefore, a number of determination results (defect color determination results) equal to the number of times this defect analysis process is repeated are obtained. Then, as mentioned above, in step S165 of Figure 15, the color that was determined to be the most frequently as a defective color in step S1644 is determined to be the final defective color.

[0096] <6.4 Correction Processing> Figure 20 is a flowchart showing the detailed procedure of the correction process performed by the first correction unit 120. After the correction process starts, it is first determined whether or not defect information (information on the location of candidate defect nozzles and information on defect colors) exists for the page to be processed (step S171). If defect information exists, the process proceeds to step S172; if no defect information exists, the correction process ends.

[0097] In step S172, with respect to the page to be processed, the areas corresponding to the candidate locations (candidate locations of the defective nozzles) extracted in the defective nozzle location extraction process, where the density value of the defective color satisfies predetermined conditions, are set as the processing area (step S172). Specifically, the processing area is set in a region where the defective color ink is applied at a density that allows it to be determined from the corrected image whether or not the correction effect of eliminating the effect of nozzle ejection defects has been achieved.

[0098] Next, a determination is made (step S173) as to whether the size of each sub-region obtained by dividing the processing target region set in step S172 into N regions (number of candidate defective nozzles) is greater than or equal to the size of the region required for defect detection by the image inspection computer 320 (hereinafter referred to as the "minimum inspectable region"). As a result, if the size of each sub-region is greater than or equal to the size of the minimum inspectable region, the process proceeds to step S174; otherwise, the process proceeds to step S175. Note that the size of the minimum inspectable region depends on the capabilities of the image inspection computer 320.

[0099] In step S174, the N candidate defective nozzles are associated with the N subregions obtained by dividing the processing area into N regions. In this way, a one-to-one correspondence is established between the N candidate defective nozzles and the N subregions.

[0100] In step S175, it is determined whether the processing area can be divided into M sub-areas (where M is an integer less than N) such that the size of each sub-area after division is greater than or equal to the size of the minimum inspectable area. If division is possible, the process proceeds to step S176; otherwise, this correction process ends.

[0101] In Step S176, M nozzles out of the N nozzles considered to be defective nozzle candidates are associated with M sub-regions obtained by dividing the processing area into M regions.

[0102] After step S174 or step S176 is completed, defect correction processing is performed on the user image (step S177). In this case, if N defective nozzle candidates are associated with N subregions in step S174, then for each of the N subregions, different nozzles from among the N nozzle candidates are set as the nozzles to be corrected, and defect correction processing is performed. On the other hand, if M defective nozzle candidates are associated with M subregions in step S176, then for each of the M subregions, different nozzles from among the M nozzle candidates are set as the nozzles to be corrected, and defect correction processing is performed.

[0103] Next, correction position information (information indicating which nozzle was set as the target nozzle for correction and at which position the defect correction process was performed) is transmitted from the print control device 100 to the image inspection computer 320 (step S178). After that, unnecessary defect information is deleted (step S179). This completes the correction process by the first correction unit 120.

[0104] In this embodiment, the mapping step is implemented by steps S173 to S176. In this regard, in steps S173 to S176, the mapping unit 122 maps N nozzles to N sub-regions obtained by dividing the processing area within one page into N sub-regions if the size of each sub-region is greater than or equal to the size of the minimum inspectable area. If the size of each sub-region is less than the size of the minimum inspectable area when the processing area within one page is divided into N sub-regions, the mapping unit 122 divides the processing area within one page into M sub-regions (M is an integer less than N) within a range where the size of each sub-region after division is greater than or equal to the size of the minimum inspectable area, and maps M nozzles out of the N nozzles to the M sub-regions.

[0105] The following describes the mapping (mapping between a sub-region and a nozzle) performed in this correction process. Here, we assume that four nozzles have been extracted as candidate defective nozzles, and these four nozzles will be referred to as "Nozzle 1," "Nozzle 2," "Nozzle 3," and "Nozzle 4."

[0106] If a streak defect 60 shown in Figure 6 occurs in an image on one page, and the area corresponding to the streak defect 60 is set as the processing area in step S172, then assume that the size of each sub-region (sub-regions 61 to 64 in Figure 7) obtained by dividing the processing area into four regions is greater than or equal to the size of the minimum inspectable area. In this case, for example, sub-region 61 is associated with the first nozzle, sub-region 62 with the second nozzle, sub-region 63 with the third nozzle, and sub-region 64 with the fourth nozzle. Then, in step S177, for sub-region 61, the first nozzle is set as the target nozzle for correction and the defect correction process is performed; for sub-region 62, the second nozzle is set as the target nozzle for correction and the defect correction process is performed; for sub-region 63, the third nozzle is set as the target nozzle for correction and the defect correction process is performed; and for sub-region 64, the fourth nozzle is set as the target nozzle for correction and the defect correction process is performed.

[0107] As shown in Figure 21, a streak defect labeled 81 and a streak defect labeled 82 occur within an image on one page, and the areas corresponding to these streak defects 81 and 82 are set as the processing area in step S172. Assume that the size of each sub-region (sub-regions 811, 812, 821, and 822 in Figure 21) obtained by dividing the processing area into four regions is greater than or equal to the size of the minimum inspectable area. In this case, for example, sub-region 811 is associated with the first nozzle, sub-region 812 with the second nozzle, sub-region 821 with the third nozzle, and sub-region 822 with the fourth nozzle. Then, in step S177 above, defect correction processing is performed for partial region 811 with the first nozzle set as the nozzle to be corrected, for partial region 812 with the second nozzle set as the nozzle to be corrected, for partial region 821 with the third nozzle set as the nozzle to be corrected, and for partial region 822 with the fourth nozzle set as the nozzle to be corrected.

[0108] As shown in Figure 22, assume that a streak defect labeled 83 occurs on the first page, and streak defects labeled 84 and 85 occur on the second page. When the areas corresponding to these streak defects 83, 84, and 85 are set as processing areas in step S172, assume that the size of each sub-area (sub-areas 831 and 832 in Figure 22) obtained by dividing the processing area corresponding to streak defect 83 into two areas, the size of the processing area 841 corresponding to streak defect 84, and the size of the processing area 851 corresponding to streak defect 85 are greater than or equal to the size of the minimum inspectable area. However, assume that the size of each sub-area obtained by dividing the processing area corresponding to streak defect 83 into three areas, the size of each sub-area obtained by dividing the processing area 841 corresponding to streak defect 84 into two areas, and the size of each sub-area obtained by dividing the processing area 851 corresponding to streak defect 85 into two areas are less than the size of the minimum inspectable area. At this time, for example, partial region 831 is associated with the first nozzle, partial region 832 is associated with the second nozzle, processing area 841 is associated with the third nozzle, and processing area 851 is associated with the fourth nozzle. Then, in step S177 above, for partial region 831, the first nozzle is set as the nozzle to be corrected and defect correction processing is performed; for partial region 832, the second nozzle is set as the nozzle to be corrected and defect correction processing is performed; for processing area 841, the third nozzle is set as the nozzle to be corrected and defect correction processing is performed; and for processing area 851, the fourth nozzle is set as the nozzle to be corrected and defect correction processing is performed. In this example, processing areas 841 and 851 are treated as partial regions.

[0109] As shown in Figure 23, assume that a streak defect labeled 86 occurs on the first page, a streak defect labeled 87 occurs on the second page, a streak defect labeled 88 occurs on the third page, and a streak defect labeled 89 occurs on the fourth page. Then, when the areas corresponding to these streak defects 86, 87, 88, and 89 are set as processing target areas in step S172, assume that the size of the processing target area 861 corresponding to streak defect 86, the size of the processing target area 871 corresponding to streak defect 87, the size of the processing target area 881 corresponding to streak defect 88, and the size of the processing target area 891 corresponding to streak defect 89 are greater than or equal to the size of the minimum inspectable area. It is assumed that the size of each sub-region obtained by dividing the processing area 861 corresponding to streak defect 86 into two regions, the size of each sub-region obtained by dividing the processing area 871 corresponding to streak defect 87 into two regions, the size of each sub-region obtained by dividing the processing area 881 corresponding to streak defect 88 into two regions, and the size of each sub-region obtained by dividing the processing area 891 corresponding to streak defect 89 into two regions are less than the size of the minimum inspectable area. In this case, for example, the processing area 861 is associated with the first nozzle, the processing area 871 is associated with the second nozzle, the processing area 881 is associated with the third nozzle, and the processing area 891 is associated with the fourth nozzle. Then, in step S177 above, defect correction processing is performed on the processing area 861 with the first nozzle set as the correction target nozzle, on the processing area 871 with the second nozzle set as the correction target nozzle, on the processing area 881 with the third nozzle set as the correction target nozzle, and on the processing area 891 with the fourth nozzle set as the correction target nozzle. In this example, processing areas 861, 871, 881, and 891 are treated as partial areas.

[0110] In the examples shown in Figures 22 and 23, the user image is an image spanning two or more pages, and the mapping unit 122 distributes four sub-regions across multiple pages and maps the four nozzles, which are candidates for defective nozzles, to these four sub-regions.

[0111] <6.5 Defective Nozzle Identification Process> Figure 24 is a flowchart showing the detailed procedure for the defective nozzle identification process. After the defective nozzle identification process begins, it is first determined whether the given imaging page (imaging image) is a target for processing (step S211). If the imaging page is a target for processing, the process proceeds to step S212; if the imaging page is not a target for processing, the defective nozzle identification process ends. For example, if the imaging page does not contain any defects, it is determined that the imaging page is not a target for processing.

[0112] Next, the correction position information transmitted from the print control device 100 to the image inspection computer 320 in step S178 of Figure 20 is read (step S212). Then, the process of determining whether the nozzle to be judged is a defective nozzle is repeated for the number of nozzles corresponding to the correction position information. In step S213, based on the correction position information and the result of the inspection process performed in step S20 of Figure 10, the nozzle to be judged is set as a target nozzle for correction and it is determined whether there is a defect at the position where defect correction processing was performed (correction position). If a defect is found at the correction position, the process proceeds to step S214; if there is no defect at the correction position, the process proceeds to step S215. In step S214, it is determined that the nozzle to be judged is not a defective nozzle. In step S215, it is determined that the nozzle to be judged is a defective nozzle.

[0113] In the example shown in Figures 6 to 8, the process for determining whether the nozzle to be judged is a defective nozzle (the loop process in Figure 24) is repeated four times. Since there is a defect in sub-region 61 (see Figure 8), the first nozzle, which was set as the nozzle to be corrected for sub-region 61 during the defect correction process, is determined not to be a defective nozzle. Similarly, since there are defects in sub-regions 62 and 64, the second and fourth nozzles are determined not to be defective nozzles. In contrast, since there is no defect in sub-region 63, the third nozzle, which was set as the nozzle to be corrected for sub-region 63 during the defect correction process, is determined to be a defective nozzle. In this way, the nozzles associated with the sub-regions in which no defects were detected by the inspection process based on the corrected image are identified as defective nozzles.

[0114] After determining whether all nozzles corresponding to the correction position information are defective nozzles, the processed image data (user image data transmitted from the print control device 100 to the image inspection computer 320) and the correction position information are deleted (step S216). This completes the defective nozzle identification process.

[0115] <7. Effects> According to this embodiment, based on the location of defects detected in an inspection process using an image obtained by capturing a user image (printed image), N candidate locations for the defective nozzle are extracted. Then, only the N nozzles corresponding to each of these N candidate locations are sequentially set as the nozzles to be corrected, and the defect correction process is performed. Subsequently, the defective nozzle is identified based on the location of defects detected in an inspection process using an image obtained by capturing the image after the defect correction process (corrected image). As described above, only N nozzles (for example, 4) corresponding to each of the N candidate locations are set as the nozzles to be corrected during the defect correction process from among the multiple nozzles provided in the printing system. Furthermore, there is no need to print a special image such as a nozzle check pattern. Thus, according to this embodiment, a printing system and a defective nozzle detection method that can efficiently detect defective nozzles from a large number of nozzles are realized.

[0116] <8. Variation> In the above embodiment, an inkjet printing apparatus 10 that performs color printing was used. However, the present invention is not limited thereto, and an inkjet printing apparatus that performs monochrome printing may also be used. In this case, the defective color identification unit 324 (see Figure 9) in the image inspection computer 320 becomes unnecessary, and a defective nozzle is identified from the multiple nozzles contained in the print head that ejects black ink without the need for a process to identify the defective color.

[0117] In the above embodiment, the present invention was described in which it is applied to an inkjet printing apparatus 10 equipped with four print heads (ink ejection units) 25K, 25C, 25M, and 25Y for black (K), cyan (C), magenta (M), and yellow (Y). However, the present invention is not limited thereto, and even when the present invention is applied to a printing apparatus that performs printing using five or more print heads, defective colors can be identified by performing a process similar to the defective color identification process described above.

[0118] Furthermore, in the above embodiment, an inkjet printing apparatus 10 using water-based ink was employed. However, the present invention is not limited thereto, and an inkjet printing apparatus using UV ink (ultraviolet-curing ink), such as an inkjet printing apparatus for label printing, may also be employed. In this case, the printing mechanism 201 (see Figure 2) is provided with an ultraviolet irradiation unit that cures the UV ink on the printing paper PA by ultraviolet irradiation, instead of a drying unit 206. [Explanation of Symbols]

[0119] 10… Inkjet printing equipment 25...Print head 100…Printing control device 110...Original image storage unit 120...First correction unit 122... Correspondence section 130...Second correction unit 200... Printing machine body 205…Printing Department 251... Head Module 252… Nozzle 300…Image inspection device 310... Imaging unit 320…Image inspection computer 321... Correct Answer Data Creation Department 322...Defect detection unit 323...Defect nozzle location candidate extraction unit 324...Defect color identification section 325... Defect nozzle identification part

Claims

1. A printing unit having multiple nozzles, which prints on a printing medium by ejecting ink from the multiple nozzles, An imaging unit that performs imaging of the printed image printed on the printing medium by the printing unit, A defect detection unit performs inspection processing to detect defects contained in the captured image obtained by capturing the printed image by the imaging unit, A defective nozzle position candidate extraction unit extracts N (where N is an integer of 2 or more) candidate positions from the positions of the plurality of nozzles as candidate positions for a defective nozzle having an ejection defect, based on the position of the defect detected by the inspection process in the captured image. A defect correction unit generates a corrected image by performing a defect correction process on a user image, which is the image to be printed, to remove the effects of nozzle ejection defects set for the nozzle to be corrected. A defective nozzle identification unit identifies the defective nozzle from the N nozzles corresponding to each of the N candidate positions. Equipped with, After printing by the printing unit, capturing the printed image by the imaging unit, and the inspection process by the defect detection unit are performed based on the user image, the N candidate positions of the defective nozzle are extracted by the defect nozzle position candidate extraction unit. The defect correction unit performs the defect correction process by sequentially setting only the N nozzles out of the plurality of nozzles as the nozzles to be corrected, one by one. After the defect correction unit generates the corrected image, the printing unit prints the corrected image, the imaging unit captures the printed image, and the defect detection unit performs the inspection process based on the corrected image. The defective nozzle identification unit identifies the defective nozzle based on the position of the defect in the captured image, which was detected by the inspection process performed based on the corrected image. The defect correction unit includes a mapping unit that maps the N nozzles to N partial regions included in the user image, Each of the N subregions includes the location of a defect detected by the inspection process performed based on the user image, The defect correction unit performs the defect correction process by setting different nozzles from the N nozzles to be corrected as the target nozzles for each of the N subregions. The printing system is characterized in that the defective nozzle identification unit identifies a nozzle as the defective nozzle that corresponds to a partial region among the N partial regions in which no defects were detected by the inspection process performed based on the corrected image.

2. The printing unit includes a plurality of ink ejection units that eject ink of different colors from nozzles. The system further includes a defective color identification unit that identifies which of the plurality of ink ejection units ejects ink of which color the defective nozzle is located in. The printing system according to claim 1, characterized in that when the defect correction unit performs the defect correction process, it sets a nozzle that ejects ink of the defect color, which is the color identified by the defect color identification unit, to be the nozzle to be corrected.

3. The plurality of ink ejection units include a black ink ejection unit for ejecting black ink, a cyan ink ejection unit for ejecting cyan ink, a magenta ink ejection unit for ejecting magenta ink, and a yellow ink ejection unit for ejecting yellow ink. The printing system according to claim 2, characterized in that the defect color identification unit identifies the defect color from among the black ink, cyan ink, magenta ink, and yellow ink, excluding the ink color whose amount to be ejected at the location of the defect detected by the inspection process performed based on the user image is less than or equal to a predetermined threshold.

4. The printing system according to claim 2 or 3, characterized in that the defective color identification unit identifies the defective color based on the difference in the average values ​​of red color values, the difference in the average values ​​of green color values, and the difference in the average values ​​of blue color values ​​in the region where a defect was detected by the inspection process performed based on the user image, between the captured image based on the user image and the correct image corresponding to the user image.

5. The printing system according to any one of claims 2 to 4, characterized in that the matching unit sets a region among the regions corresponding to the N position candidates in which the density value of the defect color satisfies a predetermined condition as a processing target region, and associates the N nozzles with the N subregions in the processing target region.

6. The aforementioned correspondence unit is If, when the processing area within one page is divided into N parts, the size of each sub-area is greater than or equal to the size of the area required for defect detection by the inspection process, then the N nozzles are associated with the N sub-areas obtained by dividing the processing area within one page into N parts. The printing system according to claim 5, characterized in that, when the processing area within a page is divided into N parts, and the size of each sub-area is less than the size of the area required for defect detection by the inspection process, the processing area within a page is divided into M sub-areas (where M is an integer less than N) within a range where the size of each sub-area after division is equal to or greater than the size of the area required for defect detection by the inspection process, and M nozzles out of the N nozzles are associated with the M sub-areas.

7. The aforementioned user image is an image consisting of two or more pages, The printing system according to any one of claims 1 to 6, characterized in that the correspondence unit distributes the N sub-regions across multiple pages and associates the N nozzles with the N sub-regions.

8. The printing system according to claim 7, characterized in that the size of each of the N subregions is greater than or equal to the size of the region required for defect detection by the inspection process.

9. The printing system according to any one of claims 1 to 8, characterized in that the defective nozzle position candidate extraction unit converts the coordinates of the position of the defect detected by the inspection process performed based on the user image into coordinates on the user image, and extracts the N position candidates from the positions of the plurality of nozzles based on the converted coordinates.

10. A method for detecting a defective nozzle in a printing system, which includes a printing unit having multiple nozzles and performing printing on a printing medium by ejecting ink from the multiple nozzles, and an imaging unit that captures a printed image printed on the printing medium by the printing unit, The first printing step involves the printing unit printing a user image, A first imaging step for capturing the printed image obtained in the first printing step, A first defect detection step for detecting defects included in the first image obtained in the first imaging step, A defective nozzle position candidate extraction step is performed to extract N (where N is an integer of 2 or more) candidate positions from the positions of the plurality of nozzles as candidate positions for a defective nozzle having an ejection defect, based on the position of the defect detected in the first defect detection step in the first image capture; A defect correction step to generate a corrected image by performing a defect correction process on the user image to remove the effects of nozzle discharge defects set for the nozzle to be corrected, A second printing step in which the printing unit prints the corrected image, A second imaging step for capturing the printed image obtained in the second printing step, A second defect detection step for detecting defects included in the second image obtained in the second imaging step, A defect nozzle identification step involves identifying the defect nozzle from N nozzles corresponding to each of the N position candidates based on the position of the defect detected in the second defect detection step in the second captured image, and Includes, In the defect correction step, only the N nozzles out of the plurality of nozzles are sequentially set one by one as the nozzles to be corrected, and the defect correction process is performed. The defect correction step includes a mapping step that associates the N nozzles with N subregions included in the user image, Each of the N subregions includes the location of the defect detected in the first defect detection step, In the defect correction step, for each of the N subregions, different nozzles from the N nozzles are set as the nozzles to be corrected, and the defect correction process is performed. A method for detecting a defective nozzle, characterized in that in the defective nozzle identification step, a nozzle associated with a sub-region among the N sub-regions in which no defect was detected in the second defect detection step is identified as the defective nozzle.

11. The printing unit includes a plurality of ink ejection units that eject ink of different colors from nozzles. The process further includes a defective color identification step of identifying which of the plurality of ink ejection units ejects ink of which color the defective nozzle is contained in, The defect nozzle detection method according to claim 10, characterized in that, when the defect correction process is performed in the defect correction step, a nozzle that ejects ink of the defect color, which is the color identified in the defect color identification step, is set as the nozzle to be corrected.

12. The plurality of ink ejection units include a black ink ejection unit for ejecting black ink, a cyan ink ejection unit for ejecting cyan ink, a magenta ink ejection unit for ejecting magenta ink, and a yellow ink ejection unit for ejecting yellow ink. The defect nozzle detection method according to claim 11, characterized in that in the defect color identification step, the defect color is identified from among the black ink, cyan ink, magenta ink, and yellow ink, excluding the ink color whose amount to be ejected at the location of the defect detected in the first defect detection step is less than or equal to a predetermined threshold.

13. The defect nozzle detection method according to claim 11 or 12, characterized in that in the defect color identification step, the defect color is identified based on the difference in the average values ​​of red, green, and blue color values ​​in the region where a defect was detected in the first defect detection step, between the captured image based on the user image and the ground truth image corresponding to the user image.

14. The defect nozzle detection method according to any one of claims 11 to 13, characterized in that, in the mapping step, a region among the regions corresponding to the N position candidates in which the density value of the defect color satisfies a predetermined condition is set as the processing target region, and the N nozzles are mapped to the N subregions within the processing target region.

15. In the aforementioned correspondence step, When the processing area within one page is divided into N parts, if the size of each sub-area is greater than or equal to the size of the area required for defect detection in the second defect detection step, then the N nozzles are associated with the N sub-areas obtained by dividing the processing area within one page into N parts. The defect nozzle detection method according to claim 14, characterized in that when the processing area within one page is divided into N parts, the size of each sub-area is less than the size of the area required for defect detection in the second defect detection step, the processing area within one page is divided into M sub-areas (where M is an integer less than N) within a range where the size of each sub-area after division is equal to or greater than the size of the area required for defect detection in the second defect detection step, and M nozzles out of the N nozzles are associated with the M sub-areas.

16. The aforementioned user image is an image consisting of two or more pages, The defective nozzle detection method according to any one of claims 10 to 15, characterized in that, in the mapping step, the N subregions are distributed across multiple pages, and the N nozzles are mapped to the N subregions.

17. The defect nozzle detection method according to claim 16, characterized in that the size of each of the N subregions is greater than or equal to the size of the region required for defect detection in the second defect detection step.

18. The defective nozzle detection method according to any one of claims 10 to 17, characterized in that in the defective nozzle position candidate extraction step, the coordinates of the position of the defect detected in the first defect detection step in the first captured image are converted to coordinates on the user image, and based on the converted coordinates, the N position candidates are extracted from the positions of the plurality of nozzles.

Images