Server, report output method, and report output system

Abstract

Obtain report information via the network that allows you to select patches based on your preferences. [Solution] The server's communication unit receives colorant value information indicating the colorant values ​​of each patch, and a chart image, from the receiving device via the network, and transmits printer-related report information to the transmitting device via the network. The server's processing unit determines the appropriate range and recommended values ​​of the colorant values ​​for each patch based on the colorant value information and the chart image, and generates reliability information indicating the reliability of the recommended values. In the overall display unit, the processing unit includes appropriate range information in the overall chart to distinguish multiple appropriate range patches from the remaining multiple non-appropriate range patches. In the recommended portion display unit, the processing unit appends the reliability information to the recommended patches corresponding to the recommended values.

Description

[Technical Field]

[0001] The present invention relates to a server capable of communicating information for a printer capable of printing charts containing multiple patches, a report output method, and a report output system. [Background technology]

[0002] As a printing device, an inkjet printer is known, which ejects ink droplets from a print head onto a printing medium. When the amount of ink ejected per unit area on the printing medium is large, phenomena such as ink bleeding into the surrounding area occur, and a color saturation state occurs where the color does not change much even if the amount of ink ejected increases. Therefore, a chart containing multiple patches for selecting the maximum ink amount, which is the upper limit of the amount of ink per unit area on the printing medium, is printed on the printing medium, and the maximum ink amount is set based on the selection result of the patches. The set maximum ink amount is used for purposes such as creating a color conversion LUT (lookup table). Furthermore, as disclosed in Patent Document 1, a printing system including a cloud printing service and a printing device is also known. [Prior art documents] [Patent Documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2021-192190 [Overview of the project] [Problems that the invention aims to solve]

[0004] One approach to determine the maximum ink volume is to obtain an inference result for the optimal maximum ink volume from a server with a pre-trained model generated by machine learning. However, even if the user is not satisfied with the maximum ink volume inferred by the pre-trained model, a color conversion LUT will be generated according to that inference result. Therefore, a new mechanism is desired to allow the user to determine the maximum ink volume. [Means for solving the problem]

[0005] The server of the present invention is a server that can communicate information for a printer capable of printing a chart containing multiple patches over a network, A communication unit that receives colorant value information indicating the colorant value of each patch, and a chart image obtained by reading the chart from a receiving device via the network, and transmits report information regarding the printer to a transmitting device via the network, The system includes an overall view display unit that shows the overall view of the chart including the plurality of patches, and a processing unit that generates the report information including a recommended partial display unit that shows the recommended patch corresponding to the recommended value of the colorant among the plurality of patches, The aforementioned processing unit, Based on the aforementioned colorant value information and the chart image, the appropriate range and recommended value of the colorant value for each patch are determined, and reliability information indicating the reliability of the recommended value is generated. In the overall display unit, appropriate range information that distinguishes multiple appropriate range patches from the remaining multiple non-appropriate range patches is included in the overall view of the chart. The recommended portion display unit has an embodiment in which the reliability information is added to the recommended patch.

[0006] Furthermore, the report output method of the present invention is a report output method in which information for a printer capable of printing a chart including multiple patches is transmitted via a network by a server that can communicate over a network, A receiving step of receiving colorant value information indicating the colorant value of each patch, and a chart image obtained by reading the chart, from a receiving device via the network, A prediction step that determines the appropriate range and recommended values ​​of the colorant values ​​for each patch based on the colorant value information and the chart image, and generates reliability information indicating the reliability of the recommended values. The overall picture display unit included in the report information relating to the printer includes an overall picture processing step in which the overall picture of the chart including the plurality of patches includes appropriate range information that distinguishes the plurality of appropriate range patches that are within the appropriate range from the remaining plurality of incorrect range patches, In the recommended portion display unit to be included in the report information, a recommended portion processing step is performed to add the reliability information to the recommended patch corresponding to the recommended value among the plurality of patches, The embodiment includes a transmission step of transmitting the report information, including the overall display unit and the recommended partial display unit, to a target device via the network.

[0007] Furthermore, the report output system of the present invention is Information for a printer capable of printing charts containing multiple patches is communicated via a network to a server, A device to transmit data connected to the aforementioned network, A report output system including, The aforementioned server is A communication unit that receives colorant value information indicating the colorant value of each patch, and a chart image obtained by reading the chart from a receiving device via the network, and transmits report information regarding the printer to a transmitting device via the network, The system includes an overall view display unit that shows the overall view of the chart including the plurality of patches, and a processing unit that generates the report information including a recommended partial display unit that shows the recommended patch corresponding to the recommended value of the colorant among the plurality of patches, The aforementioned processing unit, Based on the aforementioned colorant value information and the chart image, the appropriate range and recommended value of the colorant value for each patch are determined, and reliability information indicating the reliability of the recommended value is generated. In the overall display unit, appropriate range information is included in the overall view of the chart to distinguish the multiple appropriate range patches from the remaining multiple non-appropriate range patches among the multiple patches. In the recommended section display unit, the reliability information is added to the recommended patch. The transmission target device has an aspect of receiving the report information from the server via the network and printing or displaying the received report information.

Brief Description of Drawings

[0008] [Figure 1] A block diagram schematically showing a configuration example of a cloud printing system. [Figure 2] A block diagram schematically showing a configuration example of a server and a printer included in a cloud printing system. [Figure 3] A diagram schematically showing an example of a chart on a printing medium. [Figure 4] A diagram schematically showing examples of a teacher chart image and a test chart image. [Figure 5] A diagram schematically showing an example of generating a dataset from a plurality of teacher images with different ink amounts per unit area. [Figure 6] A diagram schematically showing an example of a learned model generated by a learned model generation device and used by a support device. [Figure 7] A diagram schematically showing an example of prediction information indicating whether the ink amount per unit area of each test image is within an appropriate range of the maximum ink amount or outside the appropriate range. [Figure 8] A diagram schematically showing an example of displaying report information including an overall image display section and a recommended section display section. [Figure 9] A flowchart schematically showing an example of a learned model generation process. [Figure 10] A flowchart schematically showing an example of a report output process. [Figure 11] A flowchart schematically showing an example of a maximum ink amount prediction process. [Figure 12] A diagram schematically showing an example of generating reliability information. [Figure 13] A diagram schematically showing another example of generating reliability information. [Figure 14] A diagram schematically showing a structural example of a color conversion look-up table. [Modes for carrying out the invention]

[0009] The embodiments of the present invention will be described below. Of course, the following embodiments are merely illustrative of the present invention, and not all of the features shown in the embodiments are necessarily essential to the solution of the invention.

[0010] (1) Summary of embodiments included in the present invention: First, an overview of the embodiments included in the present invention will be described with reference to the examples shown in Figures 1 to 14. Note that the figures in this application are schematic examples, and the magnification in each direction shown in these figures may differ, and the figures may not be consistent. Of course, the elements of this embodiment are not limited to the specific examples indicated by the reference numerals. In "Overview of Embodiments Included in the Present Invention," the text in parentheses indicates supplementary explanation of the preceding word. Furthermore, in this application, the numerical range "Min~Max" means a value greater than or equal to the minimum value Min and less than or equal to the maximum value Max.

[0011] [Aspect 1] As illustrated in Figures 1 and 2, a server 100 according to one embodiment is a server 100 capable of communicating information over a network for a printer 200 capable of printing a chart (e.g., test chart CH2) containing a plurality of patches (e.g., test patch PA2), and comprises a communication unit (e.g., communication I / F 117) and a processing unit 110. The communication unit (117) receives colorant value information 160 indicating the colorant value of each patch (PA2) and a chart image (e.g., test chart image 140) obtained by reading the chart (CH2) from a receiving device via the network, and transmits report information 500 concerning the printer 200 to a transmitting device via the network. The processing unit 110 generates the report information 500 which includes an overall image display unit 501 showing the overall image of the chart (CH2) containing the plurality of patches (PA2), and a recommended partial display unit 502 showing recommended patches 520 corresponding to recommended colorant values ​​410 among the plurality of patches (PA2). The processing unit 110 determines the appropriate range of the colorant value and the recommended value 410 for each patch (PA2) based on the colorant value information 160 and the chart image (140), and generates reliability information 530 indicating the reliability of the recommended value 410. In the overall display unit 501, the processing unit 110 includes appropriate range information 515 in the overall image of the chart (CH2) that distinguishes a plurality of appropriate range patches 511 that are within the appropriate range from the remaining plurality of non-appropriate range patches 512 among the plurality of patches (PA2). In the recommended portion display unit 502, the processing unit 110 appends the reliability information 530 to the recommended patch 520.

[0012] When a user views the report information 500, they can see multiple appropriate range patches 511 within the appropriate range of colorant values ​​from the overall view of the chart (CH2) in the overall view display unit 501, and see the recommended patch 520 corresponding to the recommended colorant value 410 in the recommended portion display unit 502. Here, reliability information 530 indicating the reliability of the recommended value 410 is attached to the recommended patch 520, so the user can select the recommended patch 520 or a patch other than the recommended patch 520 (PA2) by referring to the reliability information 530. For example, the user can select the recommended patch 520 if the reliability of the recommended value 410 is high, or select an appropriate range patch 511 other than the recommended patch 520 if the reliability of the recommended value 410 is low. Therefore, the above embodiment can provide a server that can obtain report information via a network that allows the user to select a patch considering their own preferences.

[0013] Various examples can be given to the embodiments described above. The chart may be a chart for determining the maximum ink amount, which is the upper limit of the amount of ink per unit area in a print medium, or it may be a color selection chart for adjusting the print color, etc. The colorant value of the patch includes a value corresponding to the amount of ink used to print the patch. Ink is generally a liquid containing colorants such as pigments or dyes, but it can also be a powdered solid, like toner ink. The colorant value information may include the type of ink used to print the patch. The chart may be read by a scanner or by a camera, etc. Therefore, the chart image may be a scanned image or an image captured by a camera, etc. The receiving device may be a printer or any other terminal device. Similarly, the transmitting device may be a printer or any other terminal device. The appropriate range for colorant values ​​refers to the range of colorant values ​​that are predicted to be appropriate, with a certain degree of flexibility. The recommended value for colorant values ​​refers to the value predicted to be optimal for colorant values, and is included within the aforementioned appropriate range. There are various examples of appropriate range information. The processing unit may include appropriate range information in the overall chart by making multiple appropriate range patches stand out by making multiple non-appropriate range patches lighter. Alternatively, the processing unit may include appropriate range information in the overall chart by distinguishing multiple appropriate range patches from multiple non-appropriate range patches by enclosing them with lines or the like. Reliability information is illustrated by the variance σ in Figure 12. 2 The information may be obtained by statistically processing an index corresponding to the amount of ink per unit area (for example, the appropriate index P(Duty)), or it may be information based on an index that shows the performance of a trained model, such as the F1 score exemplified in Figure 13. Furthermore, the reliability information may be shown in stages by color coding or patterns to indicate the reliability of the recommended value, or it may be shown continuously by numerical values ​​or graphs. A server is also called a server computer. A server can consist of one computer or two or more computers. Of course, the above-mentioned supplementary statement also applies in the following embodiments.

[0014] [Aspect 2] As illustrated in Figure 3, the chart (CH2) may include a plurality of patches (PA2) with different ink amounts Q1 per unit area in the printing medium ME0. The colorant value of each patch (PA2) may correspond to the ink amount Q1 per unit area of ​​that patch (PA2). The appropriate range and the recommended value 410 may be applied to the maximum ink amount Qm, which is the upper limit of the ink amount Q1 per unit area. The processing unit 110 may determine the appropriate range and the recommended value 410 for the maximum ink amount Qm based on the colorant value information 160 and the chart image (140). In the above cases, you can obtain report information from the server that allows you to determine the maximum ink volume, taking your preferences into consideration.

[0015] Here, the patch used to check the maximum ink amount may include patterns such as linear images, or it may be a solid patch with a uniform recording density. Referring to Figures 2 and 3, the recording density (RD) refers to the ratio (including percentages) of the number of dots DT0 formed by ink droplets 237 to a predetermined number of pixels PX0 on the printing medium ME0. If dots of different sizes are formed, the ratio refers to the ratio converted to the largest dot (e.g., large dot). A pixel PX0 is the smallest element that constitutes an image, to which a color can be independently assigned. Figure 3 shows 25 pixels PX0, but if Nd large dots are formed for 100 pixels PX0, the recording density RD will be Nd%. The ink amount per unit area Q1 refers to the amount of ink ejected from the print head 230 to a unit area of ​​the printing medium ME0, and corresponds to the amount of ink required to form a patch PA0 of recording density RD on the printing medium ME0, and is substantially equal to the recording density RD. The above-mentioned supplementary statement also applies in the following embodiments.

[0016] [Aspect 3] As illustrated in Figures 8, 10, and 11, the processing unit 110 may generate consideration information 540 representing considerations for determining the recommended value 410 based on the colorant value information 160 and the chart image (140), and may also generate report information 500 that further includes the consideration information 540. In the above case, the user can select a patch (PA2) from the chart (CH2) by referring to the server 100's considerations for determining the recommended value 410. Therefore, in the above configuration, the user can obtain report information from the server that is suitable for patch selection.

[0017] [Aspect 4] As illustrated in Figure 1, the receiving device may be a terminal 180. The transmitting device may be a printer 200. The processing unit 110 may generate the report information 500 to be printed by the printer 200. In the above case, the report information 500, which includes the recommended patch 520, is printed on the print medium ME0 by the printer 200, so the user can check the print quality corresponding to the recommended value 410 in the report information 500. Therefore, in the above embodiment, report information suitable for patch selection can be obtained from the server.

[0018] [Aspect 5] The processing unit 110 may generate the report information 500 (see Figure 8) such that the recommended patch 520 is equal to or greater than the size of the patch (PA2) in the chart (CH2). In the above case, the recommended patch 520 corresponding to the recommended value 410 will be printed at a size greater than or equal to the size of the patch (PA2) in the chart (CH2), so the user can confirm the print quality corresponding to the recommended value 410 in the report information 500. Therefore, in the above configuration, suitable report information for patch selection can be obtained from the server.

[0019] [Aspect 6] As illustrated in Figure 8, the processing unit 110 may include the correct range information 515, in which the plurality of incorrect range patches 512 have been lightened or hidden, in the overall view of the chart (CH2) in the overall view of the chart (CH2) in the overall view display unit 501. In the overall display unit 501, if multiple non-correct range patches 512 are faint or hidden, multiple correct range patches 511 become more prominent. This allows the user to easily identify the correct range patch (PA2) among multiple patches (PA2). Therefore, in the above configuration, suitable report information for patch selection can be obtained from the server.

[0020] [Aspect 7] As illustrated in Figures 8, 10, 12, etc., the processing unit 110 may control the communication unit (117) to send warning information 550 indicating that the reliability information 530 is lower than a predetermined standard to the target device if the reliability information 530 is lower than a predetermined standard. An example of reliability information 530 being lower than a predetermined standard is the standard deviation σ illustrated in Figure 12. 2Examples include the fact that the value is greater than a predetermined threshold TS2, and that the F1 score, as illustrated in Figure 13, is less than a predetermined threshold TR1. In the above case, the user can understand that the reliability of the recommended value 410 is low, according to warning information 550. Therefore, in the above configuration, suitable report information for patch selection can be obtained from the server.

[0021] Here, the warning information may be information to be printed on a printer, information to be displayed on the target device, or information to be output as sound on the target device.

[0022] [Aspect 8] Incidentally, one aspect of the report output method is a report output method performed by a server 100 that can communicate information via a network for a printer 200 capable of printing a chart (CH2) containing multiple patches (PA2). As illustrated in Figure 10, this report output method includes the following steps. (a1) A receiving step ST1 in which colorant value information 160 indicating the colorant value of each patch (PA2) and a chart image (140) obtained by reading the chart (CH2) are received from a receiving device via the network. (a2) Prediction step ST2, which determines the appropriate range and recommended value 410 of the colorant value for each patch (PA2) based on the colorant value information 160 and the chart image (140), and generates reliability information 530 indicating the reliability of the recommended value 410. (a3) Overall picture processing step ST3, in an overall picture display unit 501 included in the report information 500 relating to the printer 200, the overall picture of the chart (CH2) including the plurality of patches (PA2) includes appropriate range information 515 that distinguishes a plurality of appropriate range patches 511 that are within the appropriate range from the remaining plurality of incorrect range patches 512. (a4) A recommended portion processing step ST4 in which, in the recommended portion display section 502 included in the report information 500, the reliability information 530 is attached to the recommended patch 520 among the plurality of patches (PA2) that corresponds to the recommended value 410. (a5) Transmission step ST6, which transmits the report information 500, including the overall display unit 501 and the recommended part display unit 502, to the transmission target device via the network.

[0023] The above embodiment can provide a report output method that allows users to obtain report information from a server that enables them to select patches based on their own preferences.

[0024] Furthermore, the above-described embodiments are applicable to a report output system including the server and one or more transmission target devices, a printing system including the server and one or more printers, a printing system including the server, one or more terminals and one or more printers, a report printing method including a report output method, a report generation program for generating report information, a computer-readable non-temporary medium on which the report generation program is recorded, and the like.

[0025] (2) Example configuration of a printing system including a trained model generation device and support device: Figure 1 schematically illustrates the configuration of a cloud printing system that provides report information 500 for printer 200 to user US1 via network NE1. The printing system 1 shown in Figure 1 includes an AI (artificial intelligence) server 101, a print server 102, one or more terminals 180, and one or more printers 200. These elements (101, 102, 180, 200) are connected to a network NE1. Network NE1 may be the Internet, which connects numerous communication devices worldwide according to a common communication specification; it may be a wide-area network within a limited range; or it may be a LAN (Local Area Network), etc. Since the AI ​​server 101 is an explainable AI server, it can also be called an XAI server. The combination of AI server 101 and print server 102 is an example of server 100. Of course, the AI ​​server 101 may be two or more computers, the print server 102 may be two or more computers, and server 100 may be a single computer.

[0026] As shown in Figure 1, user US1 can obtain report information 500 about the printer 200 by having the printer 200 form a print medium ME0 having a test chart CH2 containing multiple test patches PA2, and having the terminal 180 read the test chart CH2. The terminal 180 transmits colorant value information 160, which shows the amount of ink Q1 per unit area (see Figure 3) of each test patch PA2 in the print medium ME0, and a test chart image 140 obtained by reading the test chart CH2, to the AI ​​server 101 via the network NE1. The AI ​​server 101 receives the colorant value information 160 and the test chart image 140 from the terminal 180 via the network NE1, generates report information 500 based on the colorant value information 160 and the test chart image 140, and transmits it to the print server 102. Terminal 180 is an example of a receiving device. The AI ​​server 101 stores a trained model 300 for outputting appropriate range prediction information 400, recommended value 410 for the maximum ink amount Qm (see Figure 11), which is the upper limit of the ink amount Q1 per unit area on the printing medium ME0. The report information 500 includes an overall display unit 501 that shows the overall picture of the test chart CH2, including appropriate range information 515 determined from the prediction information 400, and a recommended part display unit 502 that adds reliability information 530 to the recommended patch 520 corresponding to the recommended value 410. The reliability information 530 indicates the reliability of the recommended value 410. The print server 102 receives the report information 500 from the AI ​​server 101, generates print data PD1 including the report information 500 to be printed by the printer 200, and sends it to the printer 200 via the network NE1. The printer 200 is an example of a transmission target device. Printer 200 receives print data PD1 from print server 102 via network NE1 and prints report information 500 onto print medium ME0.

[0027] When user US1 views the report information 500, the overall display unit 501 allows them to confirm the appropriate range of the maximum ink amount Qm from the overall view of test chart CH2, and the recommended patch 520 corresponding to the recommended value 410 can be confirmed in the recommended section display unit 502. Since reliability information 530 is attached to the recommended patch 520, user US1 can select the recommended patch 520 or select a test patch PA2 other than the recommended patch 520 by referring to the reliability information 530.

[0028] Terminal 180 includes a CPU (Central Processing Unit) 181, ROM (Read Only Memory) 182, RAM (Random Access Memory) 183, a reader 184, a storage device, an input device, a display device, a communication interface (I / F) connected to the network NE1, etc. Examples of terminal 180 include personal computers including tablet terminals, mobile phones including smartphones, and digital cameras including digital video cameras. Examples of reader 184 include a camera capable of capturing images of a chart including test chart CH2, and a scanner capable of reading a chart including test chart CH2. The reader 184 may also be an external device connected to the main body of terminal 180.

[0029] Figure 2 schematically illustrates the configuration of the server 100 and printer 200 included in the printing system 1. The server 100 shown in Figure 2 includes an AI server 101, which includes a trained model generation device 2 and part of the support device 3, and a print server 102, which includes part of the support device 3. The trained model generation device 2 and the support device 3 are devices for assisting in the determination of the maximum ink amount Qm. Since the AI ​​server 101 and the print server 102 have similar hardware configurations, their hardware configurations are described collectively as the hardware configuration of the server 100. The printer 200 shown in Figure 2 collectively refers to a reference printer 201 used for generating trained models 300, and a client printer 202 that outputs report information 500 to the user US1. Since the reference printer 201 and the client printer 202 have similar hardware configurations, their hardware configurations are described collectively as the hardware configuration of the printer 200. Figure 3 schematically illustrates a chart CH0 on the printing medium ME0. Figure 3 shows the teacher chart CH1 and test chart CH2 together as chart CH0. In Figure 3, a schematic diagram illustrating an example of the ink amount per unit area Q1 is shown within the dashed-dot box. The ink amount per unit area Q1 of patch PA0 corresponds to the colorant value of patch PA0. The printer 200 can form a printed image IM0 that includes chart CH0.

[0030] Server 100 includes a CPU 111, ROM 112, RAM 113, storage device 114, input device 115, display device 116, communication interface 117, etc. The aforementioned elements (111 to 117) are electrically connected and capable of inputting and outputting information to each other. ROM 112, RAM 113, and storage device 114 are memories, and at least ROM 112 and RAM 113 are semiconductor memories. Server 100 includes a processing unit 110, mainly composed of the CPU 111. RAM 113 is an example of a data storage unit. Input device 115 is an example of an operation unit. Display device 116 is an example of a display unit. Communication interface 117 is an example of a communication unit.

[0031] The storage device 114 of the AI ​​server 101 stores an OS (operating system), a learning program PR1, a maximum ink volume prediction program PR2, etc. (not shown). The learning program PR1 causes the AI ​​server 101, which is a computer, to function as a trained model generation device 2. To execute the learning program PR1, multiple training images 121 included in the training chart CH1 shown in Figure 3, and multiple labels LA1 associated with each of the training images 121, are stored in the RAM 113. After the execution of the learning program PR1, the trained model 300 is stored in the storage device 114. The maximum ink volume prediction program PR2 causes the AI ​​server 101 to function as a support device 3. To execute the maximum ink volume prediction program PR2, multiple test images 141 included in the test chart CH2 shown in Figure 3 are stored in the RAM 113. After the execution of the maximum ink volume prediction program PR2, prediction information 400 is stored in the RAM 113. Furthermore, since both RAM 113 and storage device 114 are memory, storage device 114 may function as an information storage unit, or RAM 113 may store the learned model 300. The storage device 114 of the print server 102 stores the color conversion LUT (lookup table) 600 shown in Figure 14. The color conversion LUT 600 shown in Figure 14 defines the correspondence between the coordinate values ​​of R (red), G (green), and B (blue) and the coordinate values ​​of C (cyan), M (magenta), Y (yellow), and K (black) for multiple grid points GD1. The variable i shown in Figure 14 is a variable that identifies each grid point GD1. Examples of storage devices 114 include non-volatile semiconductor memory such as flash memory, and magnetic storage devices such as hard disks.

[0032] Examples of input devices 115 include pointing devices, hard keys such as keyboards, and touch panels attached to the surface of display panels. Input devices 115 may also be external devices connected to the main body of server 100. Examples of display devices 116 include liquid crystal displays and organic EL displays. Display devices 116 may also be external devices connected to the main body of server 100. Communication I / F 117 is connected to network NE1 and inputs and outputs information to terminal 180 and printer 200. For example, the communication I / F 117 of AI server 101 receives colorant value information 160 and test chart image 140 from terminal 180 via network NE1. The communication I / F 117 of print server 102 sends report information 500 to printer 200 via network NE1.

[0033] The CPU 111 reads information stored in the memory device 114 into the RAM 113 as needed and performs various processing by executing the read program. The CPU 111 of the AI ​​server 101 performs processing corresponding to the functions of the trained model generation device 2 by executing the learning program PR1 read into the RAM 113. The CPU 111 of the AI ​​server 101 also performs processing corresponding to the functions of the support device 3 by executing the maximum ink amount prediction program PR2 read into the RAM 113. The CPU 111 of the print server 102 performs color conversion processing, halftone processing, print data generation processing, print data transmission processing, etc. by executing a print control program (not shown). For example, as part of the color conversion processing, the CPU 111 assigns R, G, and B to each pixel. 8 The process converts RGB data having integer values ​​greater than or equal to the number of gradations into ink quantity data according to the color conversion LUT600 in Figure 14. The ink quantity data is, for example, C, M, Y, and K for each pixel. 8It has integer values ​​greater than or equal to the number of gradations. The CPU 111 performs a process to generate dot data with a reduced number of gradations by performing a halftone process on the ink amount data. The CPU 111 also performs a process to generate print data PD1 by adding command data to the dot data as a print data generation process. The computer-readable non-temporary recording medium storing the programs (PR1, PR2, etc.) is not limited to the internal storage device of the server 100, but may be an external recording medium of the server 100. Furthermore, the number of CPUs 111 in the processing unit 110 may be one or two or more. Also, part or all of the processing unit 110 can be replaced with hardware such as a DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), PLD (Programmable Logic Device), FPGA (Field Programmable Gate Array), etc. Additionally, the trained model generation device 2 and the support device 3 may be configured as separate computers.

[0034] The printer 200 shown in Figure 2 is an inkjet printer that sprays C (cyan), M (magenta), Y (yellow), and K (black) inks, which are colorants, onto the printing medium ME0 from the print head 230. Therefore, there are four types of inks 236 with different colors, as shown in Figure 2. The printer 200 includes a controller 210, a communication I / F 220, a print head 230, a drive unit 250, etc. The printer 200 may also include a reader 260 capable of reading the chart CH0 on the printing medium ME0. A chart image (see Figure 4) including a teacher chart image 120 and a test chart image 140 may be generated by the reader 260 reading the chart CH0. Examples of the reader 260 include a scanner, an imaging device, etc. The reader 260 may also be an external device connected to the main body of the printer 200.

[0035] The controller 210 includes a CPU 211, ROM 212, RAM 213, drive signal transmission unit, etc., and controls the operation of the communication I / F 220, print head 230, drive unit 250, etc. The controller 210 controls the ejection of ink droplets 237 by the print head 230 according to the dot data contained in the print data PD1 acquired from the print server 102. The controller 210 may also control the relative movement of the print medium ME0 and the print head 230 by the drive unit 250. In this way, a print image IM0 corresponding to the print data PD1 is formed on the print medium ME0. The controller 210 can be configured using an SoC (System on a Chip) or the like. The communication interface 220 is connected to network NE1 and inputs and outputs information to server 100.

[0036] The print head 230 is equipped with a drive circuit and drive elements, and performs printing by ejecting ink droplets 237 onto the printing medium ME0 from a plurality of nozzles 234 included in a nozzle row 233. Here, a nozzle means a small hole from which ink droplets are ejected, and a nozzle row means a sequence of multiple nozzles. The print head 230 shown in Figure 2 includes a C nozzle row 23C that ejects C ink droplets 237, an M nozzle row 23M that ejects M ink droplets 237, a Y nozzle row 23Y that ejects Y ink droplets 237, and a K nozzle row 23K that ejects K ink droplets 237. The drive elements can include piezoelectric elements that apply pressure to the ink in a pressure chamber communicating with the nozzles 234, and drive elements that generate bubbles in the pressure chamber using heat to eject ink droplets 237 from the nozzles 234. For example, if the binary dot data based on the print data is "dot formation", the controller 210 outputs a drive signal to the print head 230 to eject ink droplets for dot formation. Furthermore, if the dot data is data with three or more values, the controller 210 outputs a drive signal to eject ink droplets for large dots if the dot data is "large dot formation," and outputs a drive signal to eject ink droplets for small dots if the dot data is "small dot formation." The printing medium ME0 is not particularly limited and includes paper, fabric, resin, metal, etc. The shape of the printing medium ME0 may be a cut two-dimensional shape or a roll.

[0037] As shown in Figure 3, chart CH0 on the printing medium ME0 contains multiple pattern sequences P0, each containing multiple patches PA0 with different ink ejection amounts 236. The multiple pattern sequences P0 shown in Figure 3 include primary color pattern sequences P11, P12, P13, P14, secondary color pattern sequences P21, P22, ..., and tertiary color pattern sequence P31. Primary colors are colors represented by only one type of ink, secondary colors are colors represented by two different inks, and tertiary colors are colors represented by three different inks. In each pattern sequence P0, the patches PA0 are arranged in ink quantity order QO1, which represents the order of ink quantity Q1 per unit area. Note that patch PA0 is a general term for the teacher patch PA1 included in the teacher chart CH1 and the test patch PA2 included in the test chart CH2.

[0038] In Figure 3, within the dashed-dotted box, a schematicly simplified example is shown where 5 × 5 = 25 pixels PX0 correspond to a predetermined number of pixels PX0 per unit area. Of course, the predetermined number corresponding to a unit area is not limited to 25, and a larger area may be treated as the unit area. The ink amount Q1 per unit area means the ratio (including percentages) of the number of ink droplets 237 ejected for a predetermined number of pixels PX0, and if ink droplets 237 of different sizes are ejected for pixels PX0, it means the ratio converted to the largest ink droplet. Within the dashed-dotted box in Figure 3, it is shown that the ink amount Q1 per unit area of ​​patch PA0 is (20 / 25) × 100 = 80%. Note that when a mixed color image such as a secondary color is formed, multiple types of ink droplets 237 are ejected for a single pixel PX0, so Q1 > 100% may occur. For example, the ink amount Q1 per unit area for a secondary color can be up to 200%. Server 100 is configured to receive colorant value information 160 from terminal 180 in order to identify each patch PA0 in chart CH0. In this specific example, the colorant value information 160 includes the amount of ink per unit area Q1 of each patch PA0, as well as the types of inks 236 used to print the patch PA0.

[0039] Each patch PA0 is rectangular and contains multiple solid areas PA3 and multiple line areas PA4. Figure 3 shows that each patch PA0 has four solid areas PA3, and line areas PA4 exist between the solid areas PA3. A solid area PA3 represents an area where the type of ink 236 does not change and the amount of ink Q1 per unit area is uniform. A line area PA4, where ink 236 is ejected, also represents an area where the type of ink 236 does not change and the amount of ink Q1 per unit area is uniform. For example, the primary color pattern sequence P11 contains a solid area PA3 of C and a line area PA4 of M, and the primary color pattern sequence P12 contains a solid area PA3 of M and a line area PA4 of Y. The secondary color pattern sequences P21, P22, ... contain solid areas PA3 of the secondary color and may also contain line areas PA4 of the secondary color. A line area PA4 may be an area where ink 236 is not ejected.

[0040] By observing a printing medium ME0 on which multiple patches PA0 with different ink amounts Q1 per unit area are formed, it is possible to determine the relationship between phenomena such as "breaks" in line regions PA4, "thinning" in line regions PA4, "thickening" in line regions PA4, "adjacency" of line regions PA4, ink "bleeding," ink "aggregation," and ink "overflow," and the ink amount Q1 per unit area. "Breaking" in line regions PA4 refers to the phenomenon where a part of the line region PA4 is missing. "Thinning" in line regions PA4 refers to the phenomenon where the line region PA4 becomes thinner than its original width, although it does not result in a complete break. "Thickening" in line regions PA4 refers to the phenomenon where the line region PA4 becomes thicker than its original width. "Bleeding" in ink refers to the phenomenon where the outline of the patch PA0 becomes blurred due to ink bleeding into the surrounding area. "Aggregation" in ink refers to the phenomenon where the dispersibility of ink dots decreases due to the aggregation of colorants. Ink "overflow" refers to the phenomenon where the shape of patch PA0 is distorted due to ink spilling out of the original area of ​​patch PA0. These phenomena are described in Japanese Patent Publication No. 2021-24152.

[0041] If the maximum ink amount Qm (see Figure 10), which is the upper limit of the ink amount Q1 per unit area in the printing medium ME0, is too high, the color in the dark areas of the printed image IM0 becomes saturated, resulting in a decrease in image quality. On the other hand, if the maximum ink amount Qm is too low, the color reproduction of the printed image IM0 decreases. Repeated tests revealed that the above phenomenon occurs locally within patch PA0, rather than across the entire patch PA0, and that this localized phenomenon affects the image quality of the printed image IM0. Therefore, the trained model generation device 2 in this specific example generates a trained model 300 to obtain a predicted value PV1 that indicates the probability that the ink amount Q1 per unit area of ​​each divided test image 142 is appropriate as the maximum ink amount Qm, as illustrated in Figures 4 to 6. However, if the recommended value obtained from the inference results of the trained model 300 is automatically determined as the maximum ink amount Qm, a color conversion LUT will be generated according to the inference results even if the user is not satisfied with the determined maximum ink amount Qm. Therefore, the support device 3 in this specific example provides the user with predicted information 400 indicating whether the ink amount Q1 per unit area of ​​each test image 141 is within the appropriate range for the maximum ink amount Qm or outside that appropriate range, as illustrated in Figures 7 and 8. The trained model generation device 2 and the support device 3 are positioned to support the user in determining the maximum ink amount Qm.

[0042] Figure 4 schematically illustrates the teacher chart image 120 and the test chart image 140. Figure 4 shows the teacher chart image 120 and the test chart image 140 together. The training chart image 120 is obtained by reading the training chart CH1 shown in Figure 3, for example, with a reader 260 (see Figure 2). When the reader 260 reads the training chart CH1 on the printed image IM0, it generates multiple training images 121 corresponding to multiple training patches PA1 included in the training chart CH1. Thus, multiple training images 121 are obtained by reading multiple training patches PA1 with different ink amounts Q1 per unit area. These multiple training images 121 are transmitted to the AI ​​server 101 via the network NE1. When the AI ​​server 101 receives the multiple training images 121, it stores them in the RAM 113. The AI ​​server 101 may also store the multiple training images 121 in the storage device 114. In order to generate the trained model 300 shown in Figure 6, each training image 121 is divided vertically and horizontally to form N divided training images 122.

[0043] The test chart image 140 is obtained by reading the test chart CH2 shown in Figure 3, for example, with a reader 184 (see Figure 1). When the reader 184 reads the test chart CH2 on the printed image IM0, it generates multiple test images 141 corresponding to multiple test patches PA2 contained in the test chart CH2. Thus, multiple test images 141 are obtained by reading multiple test patches PA2 with different ink amounts Q1 per unit area. These multiple test images 141 are transmitted to the AI ​​server 101 via the network NE1. When the AI ​​server 101 receives the multiple test images 141, it stores them in the RAM 113. The AI ​​server 101 may also store the multiple test images 141 in the storage device 114. To utilize the trained model 300 shown in Figure 6, each test image 141 is divided vertically and horizontally to form N divided test images 142. That is, the number of divisions of the test image 141 is the same as the number of divisions N of the training image 121. Furthermore, the number of divisions N is not particularly limited; it can be any number that allows for the detection of the aforementioned phenomenon, such as 50 to 5000 divisions.

[0044] Figure 5 schematically shows an example of generating a dataset DS1 from multiple training images 121 with different ink amounts Q1 per unit area. In the label table TA1 shown in Figure 5, "Duty" represents the ink amount Q1 per unit area. First, as shown in label table TA1, a correspondence is made between the training image 121 and label LA1 for each ink amount Q1 per unit area. In Figure 5, each label LA1 is "1" if the ink amount Q1 per unit area of ​​the corresponding training patch PA1 is appropriate as the maximum ink amount Qm, "0" if it is above appropriate, and "2" if it is below appropriate. Label LA1 can be said to indicate whether the ink amount Q1 per unit area of ​​each training patch PA1 is appropriate as the maximum ink amount Qm, above appropriate, or below appropriate. Of course, the value of label LA1 can be changed as needed. In this specific example, label table TA1 is generated for each pattern sequence P0 shown in Figures 3 and 4. Label LA1 is assigned by an observer who has viewed the training chart CH1. In other words, for each pattern sequence P0, the observer assigns the label "1" to the ink amount per unit area Q1 of the teacher patch PA1 that is judged to be appropriate as the maximum ink amount Qm, the label "0" to the ink amount per unit area Q1 of the teacher patch PA1 that is judged to be above the appropriate maximum ink amount Qm, and the label "2" to the ink amount per unit area Q1 of the teacher patch PA1 that is judged to be below the appropriate maximum ink amount Qm. In this specific example, for each pattern sequence P0, there is only one ink amount per unit area Q1 to which the label "1", meaning "appropriate", is assigned.

[0045] Next, each training image 121 is divided into N segmented training images 122, and all segmented training images 122 are associated with the label LA1 corresponding to the original training image 121. For example, the trained model generator 2 divides the training image "T1_100" with Q1=100% into N segmented training images "T1_100_1" to "T1_100_N", and associates all segmented training images "T1_100_1" to "T1_100_N" with the label "0" for Q1=100%. The same trained model generator 2 divides the training image "T1_90" with Q1=90% into N segmented training images "T1_90_1" to "T1_90_N", and associates all segmented training images "T1_90_1" to "T1_90_N" with the label "1" for Q1=90%. The trained model generator 2 divides the training image "T1_80" with Q1=80% into N segmented training images "T1_80_1" to "T1_80_N", and associates the label "2" with Q1=80% with all segmented training images "T1_80_1" to "T1_80_N". This collection of data becomes the dataset DS1, which is input to the neural network that becomes the trained model 300.

[0046] Figure 6 schematically illustrates a trained model 300 generated by the trained model generation device 2 and used by the support device 3. In this specific example, the trained model generation device 2 generates a trained model 300 for each type of printing medium ME0, and further generates a trained model 300 for each pattern sequence P0. If the output resolution of the printer 200 can be changed, the trained model generation device 2 may further generate a trained model 300 for each output resolution. The trained model generator 2 generates a trained model 300 by inputting a dataset DS1, in which all segmented training images 122 are associated with the label LA1, into a neural network. The trained model generator 2 repeatedly performs machine learning on the provisional trained model 300 so that the probability of the output being label LA1 increases for each input of segmented training images 122. For example, for each input to the provisional trained model 300, the trained model 300 calculates a feature vector for distinguishing label LA1 from each segmented training image 122, and repeatedly performs the aforementioned machine learning based on this feature vector. It can also be said that the neural network performs machine learning based on the relationship between label LA1 and multiple segmented training images 122. The resulting trained model 300 can output a predicted value PV0 indicating the probability that the segmented test image 142 corresponds to label "0", a predicted value PV1 indicating the probability that the segmented test image 142 corresponds to label "1", and a predicted value PV2 indicating the probability that the segmented test image 142 corresponds to label "2", based on the input of a segmented test image 142. The trained model 300 causes the AI ​​server 101 to function to obtain a predicted value PV1 that indicates the probability that the ink amount Q1 per unit area of ​​the segmented test image 142 is appropriate as the maximum ink amount Qm, based on the segmented test image 142.

[0047] In order to input multiple segmented test images 142 into the trained model 300, first, a test image 141 is associated with each ink amount Q1 per unit area, as shown in the test image table TA2. Next, each test image 141 is divided into N segmented test images 142. For example, the support device 3 divides the test image "T2_100" with Q=100% into N segmented test images "T2_100_1" to "T2_100_N". The same support device 3 divides the test image "T2_90" with Q=90% into N segmented test images "T2_90_1" to "T2_90_N", and divides the test image "T2_80" with Q=80% into N segmented test images "T2_80_1" to "T2_80_N". These split test images 142 are input to the trained model 300, and a predicted value PV1 output from the trained model 300 is obtained for each split test image 142.

[0048] However, since there are N predicted values ​​PV1 for each ink amount Q1 per unit area, the support device 3 calculates the appropriate index P by performing statistical processing on the N predicted values ​​PV1 for each ink amount Q1 per unit area. When averaging is performed as the statistical processing, the support device 3 calculates the appropriate index P as the arithmetic mean of the N predicted values ​​PV1. Of course, a geometric mean or the like may be calculated instead of the arithmetic mean. Alternatively, the support device 3 may arrange the N predicted values ​​PV1 in order (ascending or descending) and calculate the appropriate index P as the median of the N predicted values ​​PV1 according to this order. In either case, the appropriate index P indicates the probability that the ink amount Q1 per unit area corresponding to the test image 141 is appropriate.

[0049] Figure 7 schematically shows an example of predictive information 400 indicating whether the ink amount Q1 per unit area of ​​each test image 141 is within the appropriate range of the maximum ink amount Qm or outside that appropriate range. As shown in Figure 7, for each ink amount per unit area Q1, a calculated appropriate index P is associated, and based on the threshold TH1, the report information 500 indicates whether the ink amount per unit area Q1 is within the "appropriate range," "over" (exceeding the appropriate range), or "under" (below the appropriate range). The threshold TH1 is applied to the appropriate index P for each ink amount per unit area Q1. In the example shown in Figure 7, if the threshold TH1 is 10% and the ink amount per unit area Q1 is 75-90%, the appropriate indices P(75), P(80), P(85), and P(90) are greater than the threshold TH1, so it is within the "appropriate range." If the ink amount per unit area Q1 is 95-100%, the appropriate indices P(95) and P(100) are less than or equal to the threshold TH1, and Q1=95-100% exceeds the appropriate range of 75-90% ink amount per unit area, so it is "over." When the ink amount Q1 per unit area is 70% or less, the appropriate index P(70) is below the threshold TH1, and Q1 ≤ 70% is below the appropriate range of 75-90% for ink amount per unit area, so it is classified as "under". The prediction information 400 shown in Figure 7 indicates whether the ink amount Q1 per unit area of ​​the test image 141 is within the appropriate range of the maximum ink amount Qm, above the appropriate range, or below the appropriate range.

[0050] Furthermore, the appropriate range of ink per unit area Q1 includes the recommended value 410 for the maximum ink amount Qm. The support device 3 determines the recommended value 410 for the maximum ink amount Qm based on the ink per unit area Q1 corresponding to each test image 141 and the appropriate index P. The recommended value 410 may also be the ink per unit area Q1 with the largest appropriate index P. Alternatively, the recommended value 410 may be the middle ink per unit area included in the top three ink per unit area Q1 when the ink per unit area Q1 are arranged in order of appropriate index P (ascending or descending). In the example shown in Figure 7, the top three ink per unit area Q1 are within the appropriate range at 80%, 85%, and 90%, and the middle Q1 = 85% is the recommended value 410.

[0051] Figure 8 schematically shows an example of the display of report information 500, including the overall picture display section 501 and the recommended section display section 502. The overall display unit 501 shows the overall view of the test chart CH2, which includes multiple test patches PA2. In the overall display unit 501, multiple display patches 510 corresponding to each of the multiple test patches PA2 are arranged. For each pattern column P0, the multiple display patches 510 include multiple appropriate range patches 511 in which the corresponding ink amount Q1 per unit area is within the appropriate range, and multiple incorrect range patches 512 that are not within the appropriate range patch 511. The multiple incorrect range patches 512 include multiple display patches that are above the appropriate range and multiple display patches that are below the appropriate range. Therefore, among the multiple incorrect range patches 512 for each pattern column P0, the incorrect range patches that are above the appropriate range patches 511 indicate that the ink amount Q1 per unit area of ​​the test image 141 is above the appropriate range of the maximum ink amount Qm. For each pattern sequence P0, any of the multiple incorrect range patches 512 that are below the correct range patch 511 indicate that the ink amount Q1 per unit area of ​​the test image 141 is below the correct range for the maximum ink amount Qm. In the overall display unit 501, the test chart CH2 includes correct range information 515 that distinguishes the multiple correct range patches 511 that are within the correct range from the remaining multiple incorrect range patches 512 among the multiple test patches PA2.

[0052] For example, the AI ​​server 101 includes appropriate range information 515 in the overall view of the test chart CH2 in the overall view display unit 501, as shown in Figure 8, which makes multiple appropriate range patches 511 stand out by hiding multiple incorrect range patches 512. Hiding the incorrect range patches 512 means making the solid area PA3 and line area PA4 of the incorrect range patches 512 invisible. The color used to hide the incorrect range patches 512 is not particularly limited and can be gray, black, red, etc. Alternatively, the AI ​​server 101 may also include appropriate range information 515 in the overall view of the test chart CH2 in the overall view display unit 501, which makes multiple appropriate range patches 511 stand out by making multiple incorrect range patches 512 lighter. Making the incorrect range patches 512 lighter means that the solid area PA3 and line area PA4 of the incorrect range patches 512 are visible. The color used to lighten the incorrect range patches 512 is not particularly limited and can be gray, black, red, etc. In both cases, the appropriate range information 515 can be described as information that distinguishes multiple appropriate range patches 511 from multiple incorrect range patches 512 as prediction information 400.

[0053] Furthermore, in addition to the prediction information 400, the AI ​​server 101 may also include a recommended patch 520 showing the recommended value 410 shown in Figure 7 in the overall display unit 501. The overall display unit 501 shown in Figure 8 has a recommended patch 520 for each pattern sequence P0. Each recommended patch 520 shown in Figure 8 is surrounded by a thick line to make it stand out. The recommended patch 520 is an example of recommended information. Note that since the recommended patch 520 is displayed in the recommended partial display unit 502, the recommended patch 520 in the overall display unit 501 may be made lighter or hidden by overlaying a conspicuous color or the like.

[0054] The recommended section display 502 shows the recommended patch 520 corresponding to the recommended value 410 for the maximum ink amount Qm among multiple test patches PA2. In the recommended section display 502, the recommended patch 520 for each pattern sequence P0 is placed. The AI ​​server 101 generates report information 500 so that the size of the recommended patch 520 is greater than or equal to the size of each test patch PA2 in the test chart CH2 (see Figure 3). Figure 8 shows that there are six recommended patches 520 for primary colors, indicated as "monochromatic," three recommended patches 520 for secondary colors, and one recommended patch 520 for tertiary colors. In the recommended section display 502, reliability information 530 indicating the reliability of the recommended value 410 is attached to each recommended patch 520. The reliability information 530 can be generated by methods such as those shown in Figures 12 and 13, although this will be described in more detail later. The reliability information 530 shown in Figure 8 is indicated below each recommended patch 520 by a colored figure 531 indicating high reliability of the recommended value 410, a colored figure 532 indicating medium reliability of the recommended value 410, and a colored figure 533 indicating low reliability of the recommended value 410. The report information 500 shown in Figure 8 also includes explanatory fields for these figures 531-533. The display position of the reliability information 530 may be above, to the left of, or to the right of each recommended patch 520, in addition to below each recommended patch 520. The figures 531-533 are not limited to rectangles; they may also be circles, triangles, stars, etc., or icons. Furthermore, the reliability information 530 may also include textual information such as high, medium, or low, in addition to color coding.

[0055] In the recommended section display unit 502 shown in Figure 8, a warning information 550 may be added to recommended patches 520 with low reliability of the recommended value 410. The warning information 550 indicates that the reliability information 530 is lower than a predetermined standard. The warning information 550 may be displayed not only above the recommended patch 520, but also below, to the left of, or to the right of the recommended patch 520. In addition, the warning information 550 may be displayed in a part of the report information 500 other than the recommended section display unit 502.

[0056] Report information 500 may include discussion information 540, which represents the reasoning behind the decision for the recommended value 410. Figure 8 shows an example of discussion information 540 based on confidence information 530 for a recommended value of 70%, corresponding to the recommended patch 520 in the fifth column of monochrome.

[0057] (3) Specific examples of the process of generating a trained model: Figure 9 schematically illustrates the trained model generation process performed by the trained model generation device 2. The trained model generation process in steps S102 to S110 will be explained below, with reference to Figures 1 to 6. Note that the term "step" may be omitted, and the step number may be indicated in parentheses. The main component of the pre-trained model generation process is the processing unit 110 of the AI ​​server 101, as shown in Figures 1 and 2. The pre-trained model generation process begins when the AI ​​server 101 receives an instruction from the terminal 180 or input device 115 to generate a pre-trained model 300.

[0058] When the trained model generation process starts, the processing unit 110 controls the formation of a teacher chart CH1 on the printing medium ME0 as shown in Figure 3 (S102). As described above, the teacher chart CH1 includes multiple teacher patches PA1 with different ink amounts Q1 per unit area. For example, the storage device 114 stores teacher chart print data to print the teacher chart CH1 on the reference printer 201, and the processing unit 110 sends the teacher chart print data to the reference printer 201, thereby forming the teacher chart CH1 on the printing medium ME0. Note that if the teacher chart CH1 is already prepared, the process in S102 may be omitted.

[0059] Next, the processing unit 110 causes the reader 260 to read the teacher chart CH1 on the printing medium ME0, obtains the generated teacher chart image 120 (see Figure 4), and stores the teacher chart image 120 in the RAM 113 (S104). The teacher chart image 120 includes multiple teacher images 121 corresponding to multiple teacher patches PA1 with different ink amounts Q1 per unit area.

[0060] Next, the processing unit 110 assigns a label LA1 to each teacher patch PA1, associating the label LA1 with each teacher image 121 as shown in the label table TA1 in Figure 5 (S106). As described above, the label LA1 indicates whether the ink amount Q1 per unit area of ​​each teacher patch PA1 is appropriate as the maximum ink amount Qm, above appropriate, or below appropriate. The process of assigning the label LA1 may also be a process of receiving a numerical value for the label LA1 for each teacher patch PA1 via the input device 115. In this case, the observer of the teacher chart CH1 should input "1" if the ink amount Q1 per unit area of ​​the teacher patch PA1 is appropriate as the maximum ink amount Qm, input "0" if it is above appropriate, and input "2" if it is below appropriate. Once the numerical value for the label LA1 is input, the processing unit 110 generates the label table TA1 for each pattern column P0 by associating the teacher image 121 with the numerical value of the label LA1 for each ink amount Q1 per unit area.

[0061] Next, the processing unit 110 generates a dataset DS1 as shown in Figure 5 (S108). At this time, the processing unit 110 divides each training image 121 into N segmented training images 122 and associates the label LA1 corresponding to the original training image 121 with all segmented training images 122. As a result, a dataset DS1 is generated in which the label LA1 is associated with each segmented training image 122 for each pattern sequence P0.

[0062] Finally, the processing unit 110 performs machine learning using the dataset DS1 as input to generate a trained model 300 (S110). As shown in Figure 6, the trained model 300 causes the server 100 to function to obtain predicted values ​​PV0, PV1, and PV2 that indicate the probability that the divided test image 142 corresponds to label LA1, based on the input of the divided test image 142. It can be said that the processing unit 110 generates the aforementioned trained model 300 through machine learning based on the relationship between label LA1 and the multiple divided training images 122.

[0063] (4) Specific examples of report output processing: Figure 10 schematically illustrates the report output processing performed by the support device 3. Here, S202 corresponds to the receiving process ST1. S204 corresponds to the prediction process ST2. S206 corresponds to the overall picture processing process ST3. S208 corresponds to the recommended partial processing process ST4. S210 corresponds to the consideration information addition process ST5. S212 corresponds to the transmission process ST6. Figure 11 schematically illustrates the prediction processing of the maximum ink amount Qm performed in S204 of Figure 10. The report output processing from S202 to S212 will be explained below with reference to Figures 1 to 8. The main components of the report output processing are the AI ​​server 101 and the processing unit 110 of the print server 102. The report output processing begins when the AI ​​server 101 receives an instruction from the terminal 180 to determine the maximum ink amount Qm.

[0064] The terminal 180 shown in Figure 1 causes the printer 200 to print a test chart CH2 as shown in Figure 3. As described above, the test chart CH2 includes multiple test patches PA2 with different ink amounts Q1 per unit area. For example, the terminal 180 stores test chart print data to print the test chart CH2 on the printer 200, and the terminal 180 transmits the test chart print data to the printer 200 directly or via the network NE1, thereby forming the test chart CH2 on the printing medium ME0. When terminal 180 reads test chart CH2 and sends the obtained test chart image 140 and colorant value information 160 indicating the amount of ink Q1 per unit area of ​​each test patch PA2 to AI server 101 via network NE1, the processing unit 110 of AI server 101 starts the report output process.

[0065] When the report output process starts, the processing unit 110 of the AI ​​server 101 receives the test chart image 140 and colorant value information 160 from the terminal 180 via the network NE1 and stores them in the RAM 113 (S202). The test chart image 140 includes multiple test images 141 corresponding to multiple test patches PA2 with different ink amounts Q1 per unit area. Next, the processing unit 110 of the AI ​​server 101 performs prediction processing for the maximum ink amount Qm (see Figure 11) (S204).

[0066] When the prediction process shown in Figure 11 begins, the processing unit 110 obtains each test image 141 associated with the ink amount per unit area Q1 from the test chart image 140 based on the test chart image 140 and the colorant value information 160 (S302). The colorant value information 160 indicates the ink amount per unit area Q1 and the type of ink 236 for each test patch PA2 in the printing medium ME0. The processing unit 110 then extracts each test image 141 from the test chart image 140 and associates the ink amount per unit area Q1 and the type of ink 236 with each test image 141 according to the colorant value information 160. Next, the processing unit 110 obtains N divided test images 142 by dividing each test image 141 into N parts (S304).

[0067] Next, the processing unit 110 runs the trained model 300 with each divided test image 142 as input to obtain a predicted value PV1 (see Figure 6) for label "1", which represents the appropriate range (S306). In S306, N predicted values ​​PV1 are obtained for each divided test image 142. In S306, the processing unit 110 may also obtain a predicted value PV0 for label "0", which represents being above the appropriate range, or a predicted value PV2 for label "2", which represents being below the appropriate range. Next, the processing unit 110 calculates an appropriate index P as shown in Figure 7 (S308) by performing statistical processing on the N predicted values ​​PV1 obtained by running the trained model 300 for each test image 141. For example, the processing unit 110 calculates the appropriate index P as the arithmetic mean of the N predicted values ​​PV1 for each test image 141. As described above, the appropriate index P indicates the probability that the amount of ink per unit area Q1 corresponding to the test image 141 is appropriate. Note that N predicted values ​​PV0 may be added to the calculation of the appropriate index P, or N predicted values ​​PV2 may be added to the calculation of the appropriate index P.

[0068] Next, the processing unit 110 classifies the ink amount per unit area Q1 in which the appropriate index P exceeds the threshold TH1 into the "appropriate range" (S310). In the example shown in Figure 7, the appropriate indices P(75), P(80), P(85), and P(90) are greater than the threshold TH1, so Q1 = 75 to 90% is classified into the "appropriate range". It can be said that the processing unit 110 determines that the ink amount per unit area Q1 in which the calculated appropriate index P exceeds the threshold TH1 is within the appropriate range. In S310, the processing unit 110 may also classify the ink amount per unit area Q1 in which the appropriate index P does not exceed the threshold TH1 as "over" or "under". In the example shown in Figure 7, the appropriate indices P(95) and P(100), for which Q1 > 90%, are less than or equal to the threshold TH1, so Q1 = 95 to 100% is classified as "over". Furthermore, since the appropriate index P(70) for Q1 < 75% is less than or equal to the threshold TH1, Q1 ≤ 70% is classified as "under". The classification of S310 may be performed for each type of printing medium ME0, and further for each pattern sequence P0, or it may be performed for each output resolution. As described above, the processing unit 110 determines, based on the colorant value information 160 and the test chart image 140, whether the ink amount Q1 per unit area corresponding to each test patch PA2 is within the appropriate range of the maximum ink amount Qm. It can also be said that the processing unit 110 determines the appropriate range of the ink amount Q1 per unit area with respect to the maximum ink amount Qm.

[0069] Next, the processing unit 110 determines a recommended value 410 (see FIG. 7) of the maximum ink amount Qm from within the ink amount Q1 per unit area within the appropriate range (S312). For example, the processing unit 110 determines, as the recommended value 410, the ink amount per unit area for which the appropriateness index P is the largest within the ink amount Q1 per unit area within the appropriate range. Alternatively, the processing unit 110 may determine, as the recommended value 410, the ink amount per unit area at the center included in the top three ink amounts per unit area of the appropriateness index P within the ink amount Q1 per unit area within the appropriate range. As described above, the processing unit 110 determines the recommended value 410 from within the appropriate range for the maximum ink amount Qm based on the colorant value information 160 and the test chart image 140.

[0070] Next, the processing unit 110 generates reliability information 530 indicating the reliability of the recommended value 410 based on the colorant value information 160 and the test chart image 140 (S314).

[0071] FIGS. 12 and 13 schematically show an example of generating the reliability information 530. FIG. 12 shows an example of generating the reliability information 530 by performing statistical processing on the appropriateness index P (Duty) for each pattern column P0. The appropriateness index P (Duty) means the appropriateness index P that changes according to the ink amount Q1 per unit area indicated by "Duty" in the figure, and is determined based on the colorant value information 160 and the test chart image 140. In the example shown in FIG. 7, the appropriateness index P (Duty) corresponds to P(100) = 2.5%, P(95) = 3.0%, P(90) = 40%,.... Here, assuming that the number of appropriateness indices P (Duty) for each pattern column P0 is n, the variable for identifying each appropriateness index P (Duty) is j, and the appropriateness index P (Duty) corresponding to the variable j is Pj. For example, the processing unit 110 calculates the arithmetic mean Avg = (1 / n)ΣPj of the n appropriateness indices Pj, and the variance σ 2 =(1 / n)Σ(Pj - Avg), and then the reliability information 530 can be generated according to the threshold values TS1 and TS2. However, 0 < TS1 < TS2. For example, the processing unit 110 calculates the variance σ 2If it is smaller than the threshold value TS1, a graphic 531 of a color indicating high reliability is determined as the reliability information 530. The processing unit 110 has a variance σ 2 If it is larger than the threshold value TS2, a graphic 533 of a color indicating low reliability is determined as the reliability information 530. Incidentally, σ 2 >TS2 means that the reliability information 530 is lower than a predetermined standard. Therefore, as shown in FIG. 8, warning information 550 is appended to the recommended patch 520 corresponding to σ 2 >TS2. The processing unit 110 has a variance σ 2 If it is equal to or greater than the threshold value TS1 and equal to or less than the threshold value TS2, a graphic 532 of a color indicating medium reliability is determined as the reliability information 530.

[0072] FIG. 12 illustrates graphs 431, 432, and 433 of the appropriate index P(Duty) with respect to the ink amount Q1 per unit area. Since the peak of graph 431 is relatively steep, the variance σ 2 =s1 corresponding to graph 431 becomes relatively small. Since the peak of graph 433 is relatively gentle, the variance σ 2 =s3 corresponding to graph 433 becomes relatively large. Since the peak of graph 432 is between the peaks of graphs 431 and 433, the variance σ 2 =s2 corresponding to graph 432 is larger than variance s1 and smaller than variance s3. In the example shown in FIG. 12, since variance s1 is smaller than the threshold value TS1, the reliability information 530 appended to the recommended patch 520 corresponding to graph 431 is determined as the graphic 531 with high reliability. Since variance s2 is equal to or greater than the threshold value TS1 and equal to or less than the threshold value TS2, the reliability information 530 appended to the recommended patch 520 corresponding to graph 432 is determined as the graphic 532 with medium reliability. Since variance s3 is larger than the threshold value TS2, the reliability information 530 appended to the recommended patch 520 corresponding to graph 433 is determined as the graphic 533 with low reliability. Incidentally, the statistical processing of the appropriate index P(Duty) is not limited to the calculation process of the variance σ 2 and may be, for example, the calculation process of the standard deviation σ. It is also possible to generate the reliability information 530 based on the maximum value etc. of the appropriate index P(Duty).

[0073] FIG. 13 shows an example of generating reliability information 530 based on an index indicating the performance of the learned model 300 for each pattern sequence P0. For example, when the AI server 101 generates the learned model 300 for each pattern sequence P0, the AI server 101 calculates a precision ratio indicating the ratio of the number of samples that were actually appropriate to the number of samples predicted to be appropriate by the learned model 300, a recall ratio indicating the ratio of the number of samples predicted to be appropriate to the number of appropriate samples, and an F1 score. The F1 score is the harmonic mean of the precision ratio and the recall ratio, and is an example of an index indicating the performance of the learned model 300. Therefore, if the AI server 101 stores the F1 score for each pattern sequence P0 in the storage device 114 as the F1 score table TA3, the reliability information 530 can be generated according to the threshold values TR1 and TR2. However, 0 < TR1 < TR2. FIG. 13 shows that the F1 scores FS1 to FS10 for each pattern sequence P0 are stored in the F1 score table TA3.

[0074] For example, when the F1 score of the processing unit 110 of the AI server 101 is smaller than the threshold TR1, it determines a graphic 533 of a color indicating low reliability as the reliability information 530. Since the F1 score FS5 shown in FIG. 13 is smaller than the threshold TR1, the reliability information 530 appended to the recommended patch 520 included in the pattern column P of the "monochromatic 5th column" is determined as the graphic 533 with low reliability. Note that the fact that the F1 score is smaller than the threshold TR1 means that the reliability information 530 is lower than a predetermined standard. Therefore, as shown in FIG. 8, warning information 550 is appended to the recommended patch 520 corresponding to FS5<TR1. When the F1 score of the processing unit 110 is larger than the threshold TR2, it determines a graphic 531 of a color indicating high reliability as the reliability information 530. Since the F1 score FS3 shown in FIG. 13 is larger than the threshold TR2, the reliability information 530 appended to the recommended patch 520 included in the pattern column P of the "monochromatic 3rd column" is determined as the graphic 531 with high reliability. When the F1 score is greater than or equal to the threshold TR1 and less than or equal to the threshold TR2, the processing unit 110 determines a graphic 532 of a color indicating medium reliability as the reliability information 530. Since the F1 score FS4 shown in FIG. 13 is greater than or equal to the threshold TR1 and less than or equal to the threshold TR2, the reliability information 530 appended to the recommended patch 520 included in the pattern column P of the "monochromatic 4th column" is determined as the graphic 532 with medium reliability. Note that the index indicating the performance of the learned model 300 is not limited to the F1 score, and may be a precision ratio, a recall ratio, or the like.

[0075] Furthermore, the processing unit 110 may generate a first reliability information based on an index indicating the performance of the trained model 300, and a second reliability information based on statistical processing as shown in Figure 12, and may determine the final reliability information 530 to be the one with the lower reliability between the first and second reliability information. For example, if the reliability indicated by the first reliability information is low, a colored figure 533 indicating low reliability is determined to be the reliability information 530 regardless of the second reliability information. If the reliability indicated by the second reliability information is low, a colored figure 533 indicating low reliability is determined to be the reliability information 530 regardless of the first reliability information. If the reliability indicated by the first reliability information is moderate, and the reliability indicated by the second reliability information is moderate or high, a colored figure 532 indicating moderate reliability is determined to be the reliability information 530. If both the reliability indicated by the first reliability information and the reliability indicated by the second reliability information are high, a colored figure 531 indicating high reliability is determined to be the reliability information 530.

[0076] After generating the reliability information 530, the processing unit 110 of the AI ​​server 101 generates consideration information 540 representing considerations for determining the recommended value 410 based on the colorant value information 160 and the test chart image 140 (S316 in Figure 11), and terminates the prediction process for the maximum ink amount Qm. The consideration information 540 may be textual information expressing the reliability of the recommended patch 520, or it may be explanatory information derived from the inference process according to the trained model 300. For example, the processing unit 110 may generate textual information expressing the reliability of the recommended value corresponding to the recommended patch 520 as consideration information 540 based on the reliability information 530 generated based on the colorant value information 160 and the test chart image 140. Furthermore, the processing unit 110 may construct a trained model 300, similar to the machine learning model disclosed in Japanese Patent Application Publication No. 2022-138761, and obtain explanatory information derived from the process of determining the predicted value PV1 for the recommended patch 520 according to the trained model 300 as consideration information 540 from the trained model 300. Of course, the processing unit 110 may also generate consideration information 540 that includes the aforementioned text information and the aforementioned explanatory information.

[0077] After the prediction process of the maximum ink amount Qm, the processing unit 110 of the AI server 101 generates display data for displaying the overall image display unit 501 as shown in FIG. 8 (S206 in FIG. 10). For example, the processing unit 110 includes, as prediction information 400, appropriate range information 515 that makes a plurality of appropriate range patches 511 within the appropriate range among the plurality of test patches PA2 more prominent than the remaining plurality of non-appropriate range patches 512 in the overall image of the test chart CH2. Further, the processing unit 110 includes a recommended patch 520 indicating the recommended value 410 of the maximum ink amount Qm in the overall image display unit 501.

[0078] Next, the processing unit 110 of the AI server 101 generates display data for displaying the recommended part display unit 502 as shown in FIG. 8 (S208). For example, the processing unit 110 makes the recommended patch 520 of each pattern column P0 larger than the size of each test patch PA2 (see FIG. 3) in the test chart CH2, and attaches reliability information 530 indicating the reliability of the recommended value 410 corresponding to the lower side of each recommended patch 520. Further, the processing unit 110 attaches warning information 550 indicating that the reliability information 530 is lower than a predetermined standard above the recommended patch 520 with low reliability of the recommended value 410. In the example shown in FIG. 12, the processing unit 110 attaches warning information 550 to the recommended patch 520 corresponding to σ 2 >TS2. In the example shown in FIG. 13, the processing unit 110 attaches warning information 550 to the recommended patch 520 corresponding to FS5<TR1.

[0079] Next, the processing unit 110 of the AI server 101 adds the above-described consideration information 540 to the report information 500 including the overall image display unit 501 and the recommended part display unit 502 with warning information 550 attached as necessary (S210). The processing unit 110 of the AI server 101 performs a process of delivering the generated report information 500 to the print server 102.

[0080] Finally, the processing unit 110 of the AI ​​server 101 controls the communication interface 117 to send the report information 500, which includes the overall display unit 501, the recommended section display unit 502 with warning information 550 added as needed, and the consideration information 540, to the printer 200 via the network NE1 (S212). As a result, the report information 500 is sent from the print server 102 to the printer 200. When the printer 200 receives the report information 500 from the print server 102, it prints the report information 500 on the printing medium ME0 as shown in Figure 8.

[0081] When user US1 views the report information 500 on the printed medium ME0, the overall display unit 501 allows them to see multiple appropriate range patches 511 that are within the appropriate range for the maximum ink amount Qm from the overall view of the test chart CH2. The overall display unit 501 also allows user US1 to see the location of recommended patches 520 included in the multiple appropriate range patches 511 for each pattern column P0. In the recommended section display unit 502, user US1 can see that the recommended patch 520 corresponding to the recommended value 410 for the maximum ink amount Qm is larger than or equal to the size of each test patch PA2 (see Figure 3) in the test chart CH2. Here, reliability information 530 indicating the reliability of the recommended value 410 is attached to the recommended patch 520, warning information 550 is attached to recommended patches 520 with low reliability, and consideration information 540 regarding the determination of the recommended value 410 is also present in the report information 500. Therefore, user US1 can select a recommended patch 520 or a test patch PA2 other than the recommended patch 520 by referring to the reliability information 530, etc. For example, user US1 can select recommended patch 520 if the reliability of the recommended value 410 is high, or select appropriate range patch 511 other than recommended patch 520 if the reliability of the recommended value 410 is low. Based on the above, user US1 using the printing system 1 in this specific example can obtain report information 500 via network NE1 that allows them to select test patch PA2 according to their own preferences.

[0082] The setting of the maximum ink amount Qm may be primarily performed by the AI ​​server 101, or by the print server 102, terminal 180, or printer 200. Here, the entity that sets the maximum ink amount Qm will be called the processing entity. The processing entity receives input to select one of several test patches PA2 included in the test chart CH2, and sets the ink amount per unit area Q1 corresponding to the selected test patch PA2 to the maximum ink amount Qm. For example, in the test chart CH2 shown in Figure 3, if test patch PA2 with Q1=85% in the "C / M" pattern sequence P11 is selected, the processing entity sets the maximum ink amount Qm for C to 85%. In this case, the maximum ink amount Qm set will be different from the recommended value of 80% shown in the recommended patch 520 shown in Figure 8. Of course, if the recommended patch 520 in the "C / M" pattern sequence P11 is selected, the processing unit 110 sets the maximum ink amount Qm for C to the recommended value of 80%.

[0083] The maximum ink amount Qm is not limited to being set for each pattern sequence P0; it may also be set for primary colors collectively or for secondary colors collectively. In this case, if a test patch PA2 with Q1=80% in any of the primary color pattern sequences is selected, the processing unit 110 sets the maximum ink amount Qm for primary colors to 80%. When the maximum ink amount Qm for primary colors collectively is set, the trained model 300 for primary colors may be generated by machine learning based on the dataset DS1 containing the primary colors. When the maximum ink amount Qm for secondary colors collectively is set, the trained model 300 for secondary colors may be generated by machine learning based on the dataset DS1 containing the secondary colors.

[0084] The determined maximum ink amount Qm is used to create a color conversion LUT 600 (see Figure 14) which is referenced during the color conversion process. As shown in Figure 14, the coordinate values ​​of the ink amount data (C,M,Y,K) = (Ci,Mi,Yi,Ki) are associated with the grid point GD1 where the coordinate values ​​of the RGB data (R,G,B) are (Ri,Gi,Bi). In this case, the processing unit 110 generates a color conversion LUT 600 such that the sum of the ink amounts corresponding to the coordinate values ​​Ci, Mi, Yi, and Ki is less than or equal to the maximum ink amount Qm. When the color conversion process is performed according to the color conversion LUT 600 thus generated, the amount of ink per unit area in the printed image IM0 is limited to less than or equal to the maximum ink amount Qm.

[0085] Of course, the color conversion LUT is not limited to the color conversion LUT 600 described above. The input coordinate values ​​of the color conversion LUT may be the coordinate values ​​of C, M, and Y, or the coordinate values ​​of C, M, Y, and K, etc. The output coordinate values ​​of the color conversion LUT may be the coordinate values ​​of C, M, Y, K, and a spot color, etc. Examples of spot colors include Or (orange), Gr (green), Lc (light cyan) which is less concentrated than C, Lm (light magenta) which is less concentrated than M, Dy (dark yellow) which is more concentrated than Y, and Lk (light black) which is less concentrated than K, etc. Furthermore, the processing unit 110 may convert RGB data, etc. into ink amount data according to a color conversion LUT that may exceed the maximum ink amount Qm, and then generate print data after converting the ink amount of each pixel in the ink amount data to be less than or equal to the maximum ink amount Qm.

[0086] (5) Variations: The present invention can be modified in various ways. For example, the printer 200 may be connected to the terminal 180 without being connected to the network NE1. In this case, the server 100 may send the report information 500 to the terminal 180 via the network NE1, with the terminal 180 as the receiving device. By having the terminal 180 print the report information 500 from the printer 200, the user can select the test patch PA2 by referring to the report information 500 and considering their own preferences. Alternatively, the terminal 180 may display the report information 500 itself. In this case as well, the user can select the test patch PA2 by referring to the report information 500 and considering their own preferences. Of course, even if the printer 200 is connected to the network NE1, the server 100 may send the report information 500 to the terminal 180 via the network NE1. The server 100 may also receive the colorant value information 160 and the test chart image 140 from the printer 200 via the network NE1, with the printer 200 as the receiving device. The processes described above can be modified as needed, such as by changing the order of the steps. For example, in the report output process shown in Figure 10, it is possible to swap the processes of S206 and S208. Even if report information 500 does not include analysis information 540, or if report information 500 does not include warning information 550, users can still obtain report information via the network that allows them to select patches based on their preferences.

[0087] Furthermore, the dataset DS1 for machine learning may include elements other than the label LA1 and the segmented training images 122. When the trained model generator 2 generates a trained model 300 that combines primary or secondary colors, the color information of the solid region PA3 and the color information of the line region PA4 may be added to the dataset DS1. In this case, the support device 3 can obtain the predicted value PV1 by running the trained model 300 with the segmented test image 142, the color information of the solid region PA3, and the color information of the line region PA4 as input, and can output the predicted information 400. When the trained model generator 2 generates a trained model 300 that combines multiple types of printing media ME0, the type information of the printing media ME0 may be added to the dataset DS1. In this case, the support device 3 can obtain the predicted value PV1 by running the trained model 300 with the segmented test image 142 and the type information of the printing media ME0 as input, and can output the predicted information 400. In addition, elements such as output resolution may be added to the dataset DS1.

[0088] Patch PA0, which includes the teacher patch PA1 and the test patch PA2, may be a solid patch where there are no line regions PA4, the type of ink 236 does not change, and the amount of ink Q1 per unit area is uniform. Even in this case, phenomena such as ink "bleeding," ink "aggregation," and ink "overflow" may occur, so the trained model 300 can be used to obtain a predicted value PV1 and output predicted information 400.

[0089] In the process described above, for example, the determination of whether something "exceeds" a value can be replaced with the determination of whether it is "greater than or equal to" a value, and the determination of whether something is "less than or equal to" a value can be replaced with the determination of whether it is "less than" a value. Of course, the determination of whether something is "less than" a value can be replaced with the determination of whether it is "less than or equal to" a value, and the determination of whether something is "greater than or equal to" a value can be replaced with the determination of whether it is "greater than" a value. Such substitutions of judgments are also included in the embodiments of the present application.

[0090] (6) Conclusion: As described above, according to the present invention, in various embodiments, it is possible to provide a configuration that allows users to obtain report information via a network that enables them to select patches according to their own preferences. Of course, even in embodiments consisting only of the constituent elements of the independent claims, the basic functions and effects described above can be obtained. Furthermore, configurations obtained by substituting or changing the combinations of each configuration disclosed in the above-mentioned examples, configurations obtained by substituting or changing the combinations of each configuration disclosed in the prior art and the above-mentioned examples, etc., are also possible. The present invention also includes these configurations, etc. [Explanation of Symbols]

[0091] 1…Printing system, 2…Trained model generation device, 3…Support device, 100…Server, 101…AI server, 102…Print server, 110…Processing unit, 113…RAM, 115…Input device, 116…Display device, 117…Communication I / F, 120…Teacher chart image, 121…Teacher image, 122…Segmented teacher image, 140…Test chart image, 141…Test image, 142…Segmented test image, 160…Colorant Value information, 180... Terminal, 184... Reader, 200... Printer, 201... Reference printer, 202... Client printer, 300... Trained model, 400... Prediction information, 410... Recommended value, 500... Report information, 501... Overall display section, 502... Recommended partial display section, 510... Display patch, 511... Optimal range patch, 512... Inappropriate range patch, 515... Optimal range information, 520... Recommended patch, 530... Reliability Sexual information, 540... Consideration information, 550... Warning information, 600... Color conversion lookup table, CH0... Chart, CH1... Teacher chart, CH2... Test chart, IM0... Printed image, DS1... Dataset, LA1... Label, NE1... Network, ME0... Printing medium, P... Appropriate indicator, P0, P11~P14, P21~P22, P31... Pattern sequence, PA0... Patch, PA1... Teacher patch, PA2... Test patch, PD1... Print data, PR1... Learning program, PR2... Maximum ink amount prediction program, PV0, PV1, PV2... Predicted value, Q1... Ink amount per unit area, Qm... Maximum ink amount, QO1... Ink amount order, ST1... Receiving process, ST2... Prediction process, ST3... Overall picture processing process, ST4... Recommended partial processing process, ST5... Consideration information addition process, ST6... Transmission process, TH1, TR1, TR2, TS1, TS2... Threshold.

Claims

1. A server that can communicate information over a network for a printer capable of printing charts containing multiple patches, A communication unit that receives colorant value information indicating the colorant value of each patch, and a chart image obtained by reading the chart from a receiving device via the network, and transmits report information regarding the printer to a transmitting device via the network, The processing unit for generating the report information includes an overall view display unit that shows the overall view of the chart including the plurality of patches, and a recommended partial display unit that shows the recommended patch corresponding to the recommended value of the colorant among the plurality of patches, The aforementioned processing unit, Based on the aforementioned colorant value information and the chart image, the appropriate range and recommended value of the colorant value for each patch are determined, and reliability information indicating the reliability of the recommended value is generated. In the overall display unit, appropriate range information is included in the overall view of the chart to distinguish the multiple appropriate range patches from the remaining multiple non-appropriate range patches among the multiple patches. In the recommended section display unit, the reliability information is added to the recommended patch. server.

2. The chart includes a plurality of patches with different amounts of ink per unit area in the printing medium, The colorant value for each patch corresponds to the amount of ink per unit area of ​​that patch, The aforementioned appropriate range and recommended value apply to the maximum ink amount, which is the upper limit of the ink amount per unit area. The processing unit determines the appropriate range and the recommended value for the maximum ink amount based on the colorant value information and the chart image. The server according to claim 1.

3. The processing unit generates consideration information representing considerations for determining the recommended value based on the colorant value information and the chart image, and generates report information further including the consideration information. The server according to claim 1 or claim 2.

4. The receiving device is a terminal, The device to be transmitted is the printer, The processing unit generates the report information to be printed by the printer. The server according to claim 1 or claim 2.

5. The aforementioned report information is information to be printed by the printer, The processing unit generates the report information such that the recommended patch is equal to or greater than the size of the patch in the chart. The server according to claim 1 or claim 2.

6. The processing unit includes in the overall image display unit the correct range information, which has been thinned or hidden, the multiple incorrect range patches, in the overall image of the chart. The server according to claim 1 or claim 2.

7. The processing unit controls the communication unit to send a warning message to the target device indicating that the reliability information is lower than a predetermined standard if the reliability information is lower than a predetermined standard. The server according to claim 1 or claim 2.

8. A report output method for a printer capable of printing charts containing multiple patches, where information is provided by a server that can communicate over a network, A receiving step of receiving colorant value information indicating the colorant value of each patch, and a chart image obtained by reading the chart, from a receiving device via the network, A prediction step that determines the appropriate range and recommended values ​​of the colorant values ​​for each patch based on the colorant value information and the chart image, and generates reliability information indicating the reliability of the recommended values. The overall picture display unit included in the report information relating to the printer includes an overall picture processing step in which the overall picture of the chart including the plurality of patches includes appropriate range information that distinguishes the plurality of appropriate range patches that are within the appropriate range from the remaining plurality of incorrect range patches, In the recommended portion display section to be included in the report information, a recommended portion processing step is performed to add the reliability information to the recommended patch among the plurality of patches that corresponds to the recommended value, The process includes a transmission step of transmitting the report information, including the overall display unit and the recommended partial display unit, to a target device via the network. Report output method.

9. Information for a printer capable of printing charts containing multiple patches is communicated via a network to a server, A device to transmit data connected to the aforementioned network, A report output system including, The aforementioned server is A communication unit that receives colorant value information indicating the colorant value of each patch, and a chart image obtained by reading the chart from a receiving device via the network, and transmits report information regarding the printer to a transmitting device via the network, The processing unit for generating the report information includes an overall view display unit that shows the overall view of the chart including the plurality of patches, and a recommended partial display unit that shows the recommended patch corresponding to the recommended value of the colorant among the plurality of patches, The aforementioned processing unit, Based on the aforementioned colorant value information and the chart image, the appropriate range and recommended value of the colorant value for each patch are determined, and reliability information indicating the reliability of the recommended value is generated. In the overall display unit, appropriate range information is included in the overall view of the chart to distinguish the multiple appropriate range patches from the remaining multiple non-appropriate range patches among the multiple patches. In the recommended section display unit, the reliability information is added to the recommended patch. The transmitting device receives the report information from the server via the network, and prints or displays the received report information. Report output system.

Images