Printing system

Abstract

To provide a printing system capable of selecting a printing liquid suitable for a fabric.SOLUTION: A printing system 10 comprises: at least one first system 11 including a printing part 22 for printing on a fabric 99 by discharging a printing liquid on the fabric and an observation part 28 for acquiring observation data D2 by observing the fabric; and a second system 12 that is a system communicating with the at least one first system via a network, and includes an acquisition part 44 that acquires observation data before printing D3 acquired by the observation part observing the fabric before printing, a storage part 43 for storing a model M1 that is a model learned by machine learning and outputs physical property parameters of a recommended liquid as a printing liquid recommended for the fabric indicated by the observation data before printing when the observation data before printing is input, and an output part 45 for outputting recommended data D6 indicating the physical property parameters output by the model to the at least one first system.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a printing system.

Background Art

[0002] Patent Document 1 describes a printing system including a first system that performs printing on a fabric and a second system that communicates with the first system. The second system acquires information about the fabric from the first system. Based on the acquired information, the second system outputs recommended printing conditions to the first system.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In fabrics, the permeability of the printing liquid varies depending on the type. When the permeability of the printing liquid for the fabric changes, the image quality changes because the printing liquid is likely to bleed or is difficult to fix. Therefore, in the printing system described in Patent Document 1, in order to obtain a predetermined image quality, appropriate printing conditions are presented to the user according to the type of fabric. However, depending on the user, there may be cases where they want to respond by selecting a printing liquid in order to obtain a predetermined image quality.

Means for Solving the Problems

[0005] A printing system that solves the above problems comprises at least one first system including a printing unit that prints on a fabric by discharging a printing liquid onto the fabric and an observation unit that acquires observation data by observing the fabric; and a second system that communicates with the at least one first system via a network, and includes an acquisition unit that acquires pre-printing observation data obtained by the observation unit observing the fabric before printing; a storage unit that stores a model trained by machine learning, which, upon input of the pre-printing observation data acquired by the acquisition unit, outputs physical property parameters of a recommended printing liquid that is recommended for the fabric indicated by the pre-printing observation data; and an output unit that outputs recommended data indicating the physical property parameters output by the model to the at least one first system. [Brief explanation of the drawing]

[0006] [Figure 1] A block diagram showing one embodiment of a printing system. [Figure 2] This flowchart shows the estimation process for predicting recommended physical property parameters. [Modes for carrying out the invention]

[0007] The following describes one embodiment of the printing system with reference to the diagram. The printing system includes an inkjet printer that prints images such as characters and photographs onto fabrics such as woven or knitted materials by ejecting ink, which is an example of a printing liquid.

[0008] <Printing System> As shown in Figure 1, the printing system 10 includes at least one first system 11 and one second system 12. The printing system 10 includes, for example, multiple first systems 11 and one second system 12. The printing system 10 is configured such that at least one first system 11 and one second system 12 are communicated together via a network.

[0009] <System 1> The multiple first systems 11 all share a common configuration. Therefore, when describing the multiple first systems 11, we will refer to one first system 11 as a representative example.

[0010] The first system 11 is a system that performs printing on the fabric 99. The first system 11 performs printing on the fabric 99, for example, by applying a dye to the fabric 99. The first system 11 is owned by a user, for example. Multiple first systems 11 are each owned by multiple users, for example. A single user may own two or more of the multiple first systems 11.

[0011] The first system 11 includes a printing device 21. The printing device 21 is a device that performs printing on the fabric 99. The printing apparatus 21 includes a printing unit 22. The printing unit 22 is configured to print on the fabric 99 by discharging printing liquid. The printing unit 22 has a head 23. The head 23 has one or more nozzles 24. The head 23 discharges printing liquid from the nozzles 24 onto the fabric 99.

[0012] The printing unit 22 may have a carriage 25. The carriage 25 is equipped with a head 23. The carriage 25 scans the fabric 99. The printing device 21 is, for example, a serial printer in which the head 23 scans the fabric 99 together with the carriage 25. The printing device 21 may also be a line printer in which the head 23 can simultaneously eject printing fluid across the width of the fabric 99.

[0013] A cartridge 26 is installed in the printing device 21, for example. The cartridge 26 contains printing fluid. The cartridge 26 is detachable from the printing device 21. The cartridge 26 is installed in the printing unit 22, for example. When the cartridge 26 is installed in the printing device 21, the printing fluid contained in the cartridge 26 is supplied to the head 23.

[0014] The cartridge 26 has an IC chip 27. The IC chip 27 stores printing fluid data D1. Printing fluid data D1 is data related to the printing fluid. The IC chip 27 stores printing fluid data D1 related to the printing fluid contained in the cartridge 26.

[0015] The printing fluid data D1 includes data indicating the physical properties of the printing fluid. The physical properties of the printing fluid include at least one of the viscosity of the printing fluid, the surface tension of the printing fluid, and the contact angle of the printing fluid with respect to the fabric 99. The IC chip 27 stores the viscosity of the printing fluid, the surface tension of the printing fluid, and the contact angle of the printing fluid with respect to the fabric 99 as the printing fluid data D1. The physical properties of the printing fluid affect the penetration of the printing fluid into the fabric 99. In other words, the physical properties of the printing fluid affect the image quality printed on the fabric 99.

[0016] The first system 11 includes an observation unit 28. The observation unit 28 acquires observation data D2 by observing the fabric 99. Specifically, the observation unit 28 generates an observation image by observing the fabric 99. The observation unit 28 acquires the observation image of the fabric 99 as observation data D2. Observation data D2 is data related to the fabric 99. The observation image is data in which the fabric 99 has been digitized as an image.

[0017] The observation unit 28 is, for example, an X-ray CT scanner. In this case, the observation unit 28 observes the fabric 99 by irradiating it with X-rays. The observation unit 28 acquires a CT image of the fabric 99 as observation data D2. The observation unit 28 is not limited to an X-ray CT scanner; it may also be an electron microscope, an optical microscope, or a digital camera.

[0018] Observation data D2 includes pre-print observation data D3. Pre-print observation data D3 is data obtained by the observation unit 28 observing the fabric 99 before printing. Pre-print observation data D3 is obtained by the user having the observation unit 28 observe the fabric 99 before printing.

[0019] Observation data D2 includes post-printing observation data D4. The post-printing observation data D4 is data obtained by the observation unit 28 observing the fabric 99 after printing. The post-printing observation data D4 is obtained by having the user observe the printed fabric 99 with the observation unit 28.

[0020] The first system 11 includes a display unit 29. The display unit 29 is configured to display information. The display unit 29 displays, for example, information received from the second system 12. The display unit 29 is, for example, a display.

[0021] The first system 11 includes an operation unit 30. The operation unit 30 may be, for example, a pointing device or a touch panel. The user can input data to the first system 11 or give instructions to the first system 11 by operating the operation unit 30.

[0022] The first system 11 includes a control unit 31. The control unit 31 controls the first system 11. The control unit 31 controls the first system 11 in accordance with an instruction received from the user through the operation unit 30, for example. The control unit 31 communicates with the second system 12. The control unit 31 exchanges data with the second system 12.

[0023] The control unit 31 may be incorporated in the printing device 21 or may be an independent device separate from the printing device 21. The control unit 31 may be, for example, the CPU of the printing device 21 or a computer terminal connected to the printing device 21.

[0024] The control unit 31 may consist of one or more processors that perform various processes according to a computer program. The control unit 31 may consist of one or more dedicated hardware circuits, such as application-specific integrated circuits, that perform at least some of the various processes. The control unit 31 may consist of a circuit that includes a combination of processors and hardware circuits. The processor includes a CPU and memory such as RAM and ROM. The memory stores program code or instructions configured to cause the CPU to perform processes. The memory, i.e., computer-readable media, includes any readable media that can be accessed by a general-purpose or dedicated computer.

[0025] The control unit 31 acquires the printing fluid data D1. The control unit 31 acquires the printing fluid data D1, for example, by reading the printing fluid data D1 from the IC chip 27. The control unit 31 may also acquire the printing fluid data D1, for example, through the operation unit 30. In this case, the user inputs information written on the cartridge 26, for example, the model number of the cartridge 26, to the control unit 31 through the operation unit 30. The control unit 31 may acquire the printing fluid data D1, for example, by searching a database based on the input information.

[0026] The control unit 31 acquires observation data D2. For example, the control unit 31 acquires observation data D2 from the observation unit 28 each time the observation unit 28 generates an observation image. The control unit 31 may perform image analysis on the observation image. The control unit 31 may acquire feature quantities of the fabric 99 by performing image analysis on the observation image of the fabric 99 before printing. For example, the control unit 31 analyzes the spectrum obtained by applying a Fourier transform to the observation image. In this way, the control unit 31 performs image analysis on the observation image. The feature quantities of the fabric 99 are quantities that indicate the structure of the fabric 99, such as the thickness of the threads that make up the fabric 99, the density of the threads, the weaving method, and the knitting method. The control unit 31 may acquire the feature quantities of the fabric 99 as observation data D2. That is, observation data D2 is data that shows the observation image of the fabric 99, the information obtained based on that observation image, or both.

[0027] The control unit 31 may obtain feature quantities of the fabric 99 and evaluation values ​​indicating the image quality of the printed image on the fabric 99 by performing image analysis on the observed image of the fabric 99 after printing. The image quality is quantitatively evaluated by the evaluation values. Image quality includes blurring, sharpness, color, graininess, banding, gradation, etc. The evaluation values ​​are obtained by performing image analysis on the observed image, similar to the feature quantities of the fabric 99. In this case, the observation data D2 includes the evaluation values.

[0028] The control unit 31 acquires target image data D5. Target image data D5 is electronic data of an image to be printed on the fabric 99. The control unit 31 acquires the target image data D5 from, for example, a terminal that communicates with the first system 11, or a storage medium connected to the first system 11. The control unit 31 prints the image indicated by the acquired target image data D5 onto the fabric 99.

[0029] The control unit 31 may obtain change values ​​indicating changes in image quality by performing image analysis on the post-print observation data D4 and the target image data D5. The observation data D2 may also include change values. Based on the change values, the image quality can be evaluated accurately regardless of the image printed on the fabric 99.

[0030] The control unit 31 transmits printing fluid data D1 to the second system 12. That is, the control unit 31 transmits data including the physical property parameters of the printing fluid used for printing to the second system 12. The control unit 31 transmits observation data D2 to the second system 12. The control unit 31 transmits pre-printing observation data D3 to the second system 12. That is, the control unit 31 transmits data concerning the fabric 99 to be printed to the second system 12. The pre-printing observation data D3 is used as input data in the second system 12. The control unit 31 transmits post-printing observation data D4. That is, the control unit 31 transmits data concerning the fabric 99 after printing to the second system 12. The control unit 31 may also transmit target image data D5 to the second system 12. That is, the control unit 31 may transmit data indicating the target image to be printed to the second system 12.

[0031] The control unit 31 transmits multiple data acquired during printing to the second system 12 as a dataset DS. The dataset DS is data in which, for example, printing fluid data D1 and post-printing observation data D4 are related to each other. Specifically, the control unit 31 transmits to the second system 12 data indicating the physical property parameters of the printing fluid used for printing and data concerning the fabric 99 after printing with that printing fluid, in a related state. The dataset DS shows the change in image quality depending on the combination of fabric 99 and printing fluid. For example, in the dataset DS, if the compatibility between the fabric 99 and the printing fluid is good, the image quality will be high. The dataset DS is used as training data in the second system 12.

[0032] The dataset DS may include target image data D5. That is, the control unit 31 may transmit the printing solution data D1, the post-printing observation data D4, and the target image data D5 to the second system 12 in a state in which they are associated with each other. Specifically, the control unit 31 may transmit to the second system 12 the data indicating the physical property parameters of the printing solution used for printing, the data relating to the fabric 99 after printing with that printing solution, and the data indicating the target image to be printed in a state in which they are associated with each other.

[0033] The dataset DS may include pre-print observation data D3. That is, the control unit 31 may transmit the printing solution data D1, pre-print observation data D3, and post-print observation data D4 to the second system 12 in a state in which they are each associated. Specifically, the control unit 31 may transmit to the second system 12 data indicating the physical property parameters of the printing solution used for printing, data concerning the fabric 99 to be printed, and data concerning the fabric 99 after printing with that printing solution, in a state in which they are each associated. The control unit 31 may transmit to the second system 12 data D1, pre-print observation data D3, post-print observation data D4, and target image data D5 in a state in which they are each associated.

[0034] <System 2> The second system 12 is a system that communicates with the first system 11. The second system 12 exchanges information with the first system 11. The second system 12 exchanges information with multiple first systems 11. The second system 12 is owned by a vendor.

[0035] The second system 12 includes a server device 41. The server device 41 communicates with the control unit 31. The server device 41 includes a control circuit 42. The control circuit 42 controls the second system 12. The control circuit 42 includes, for example, a CPU. The control circuit 42, like the control unit 31, may consist of one or more processors, one or more dedicated hardware circuits, or a circuit including a combination of processors and hardware circuits.

[0036] The control circuit 42 has a storage unit 43. The storage unit 43 is a memory such as RAM and ROM. The storage unit 43 stores data received from the first system 11. The storage unit 43 stores data sets DS received from the first system 11. The storage unit 43 stores, for example, print fluid data D1, pre-print observation data D3, post-print observation data D4, and target image data D5. The storage unit 43 stores data sets DS received from each of the multiple first systems 11. Big data is formed by the accumulation of data sets DS.

[0037] The memory unit 43 stores a model M1 that has been trained by machine learning. Model M1 is a model that predicts an appropriate printing solution for the fabric 99. When input data is received, model M1 outputs data about the recommended printing solution, i.e., the recommended solution. Specifically, when pre-print observation data D3 is input, model M1 outputs the physical property parameters of the printing solution recommended for the fabric 99 indicated by the pre-print observation data D3, i.e., the physical property parameters of the recommended solution. More specifically, when pre-print observation data D3 is input, model M1 outputs the physical property parameters of the printing solution that it predicts will produce a predetermined or higher image quality for the fabric 99 indicated by the pre-print observation data D3. In other words, model M1 predicts that if a printing solution with the output physical property parameters is used for the fabric 99 indicated by the pre-print observation data D3, a predetermined or higher image quality will be produced. The physical properties parameters of the recommended solution include, as described above, at least one of the viscosity of the recommended solution, the surface tension of the recommended solution, and the contact angle of the recommended solution with respect to the fabric 99. It should be noted that, as a result of estimation, the printing solution already prepared by the user of the first system 11 may or may not match the recommended printing solution.

[0038] Model M1 is a model trained using a dataset DS stored in the memory unit 43. Model M1 is constructed, for example, by the control circuit 42 performing machine learning based on training data in which printing fluid data D1 and post-printing observation data D4 are related to each other. The training data consists of the dataset DS. By using dataset DS stored by multiple first systems 11 as training data, the output accuracy of Model M1 is improved.

[0039] The training data may include not only the printing solution data D1 and the post-printing observation data D4, but also pre-printing observation data D3. In the pre-printing observation data D3, since the fabric 99 is not yet coated with printing solution, the features of the fabric 99 are easier to extract than in the post-printing observation data D4. In particular, considering that the state of the fabric 99 may change due to the application of printing solution, using the pre-printing observation data D3 allows for more accurate extraction of the features of the fabric 99. Therefore, it is expected that the learning accuracy of model M1 will improve.

[0040] The training data may include target image data D5 in addition to the printing solution data D1 and post-printing observation data D4. In this case, model M1 can learn the degree of change between the image quality of the target image and the image quality shown in the post-printing observation data D4. In this case, model M1 can output recommended physical properties of the printing solution to minimize the change in image quality relative to the target image. For example, model M1 can output recommended physical properties of the printing solution to obtain image quality equivalent to that of the target image data D5. Furthermore, the training data may include evaluation values ​​or change values.

[0041] Model M1 is a regression model. Based on training data, Model M1 learns the conditions for image quality changes in relation to combinations of fabric features 99 and printing solution properties. The learning method is, for example, supervised learning using a neural network, also known as deep learning.

[0042] The server device 41 includes an acquisition unit 44. The acquisition unit 44 is an interface for acquiring information. The acquisition unit 44 acquires information from the first system 11. For example, the acquisition unit 44 acquires a dataset DS from the first system 11. That is, the acquisition unit 44 acquires training data from multiple first systems 11. For example, the acquisition unit 44 acquires pre-print observation data D3 from the first system 11. That is, the acquisition unit 44 acquires input data from multiple first systems 11.

[0043] The server device 41 includes an output unit 45. The output unit 45 is an interface for outputting information. The output unit 45 outputs information to, for example, the first system 11. The output unit 45 outputs recommended data D6 to the first system 11. The recommended data D6 is data relating to the printing fluid output by model M1. The recommended data D6 is, for example, data indicating the physical property parameters of the printing fluid output by model M1, i.e., the physical property parameters of the recommended fluid. The output unit 45 outputs the recommended data D6 to the first system 11, which is the source of the input data input to model M1, among the multiple first systems 11. The recommended data D6 is displayed on the display unit 29. This allows the user of the first system 11 to select an appropriate printing fluid for the fabric 99.

[0044] The recommended data D6 is not limited to the physical property parameters of the printing fluid output by model M1, but may also include information such as the model number and type of printing fluid that has those physical property parameters. For example, the control circuit 42 may select a printing fluid that has those physical property parameters based on the physical property parameters of the printing fluid output by model M1. In this case, the control circuit 42 selects a printing fluid that is close to the physical property parameters of the printing fluid output by model M1, for example, by searching a database. The output unit 45 may output information indicating the printing fluid selected by the control circuit 42 as recommended data D6 to the first system 11.

[0045] <Flowchart> Next, the prediction process, which shows how to predict the printing solution, will be described. The prediction process is performed in the printing system 10 before printing is performed on the fabric 99 in the first system 11. The prediction process is started, for example, when the user has the observation unit 28 observe the fabric 99 to be printed.

[0046] As shown in Figure 2, in step S11, the acquisition unit 44 acquires pre-print observation data D3 from the first system 11. In step S12, the control circuit 42 inputs the pre-print observation data D3 to the model M1. This yields the recommended data D6.

[0047] In step S13, the output unit 45 outputs the recommended data D6 to the first system 11. In step S14, the display unit 29 displays the recommended data D6. By checking the recommended data D6 displayed on the display unit 29, the user can select the appropriate printing solution for the fabric 99 to be printed on.

[0048] <Operation and Effects of Printing Systems> Next, the operation and effects of the above embodiment will be described. (1) The second system 12 includes a storage unit 43 that stores a model M1 which is a model trained by machine learning and which outputs the physical property parameters of a recommended printing solution, which is a recommended printing solution for the fabric 99 indicated by the pre-printing observation data D3 acquired by the acquisition unit 44, when D3 is input.

[0049] The pre-print observation data D3 contains information about the fabric 99. Therefore, according to the above configuration, based on the pre-print observation data D3, the user can be presented with recommended printing fluid properties, i.e., recommended fluid properties, for printing on the fabric 99 indicated by the pre-print observation data D3. This allows the user to select an appropriate printing fluid for the fabric 99.

[0050] (2) The physical properties parameters of the recommended solution include at least one of the viscosity of the recommended solution, the surface tension of the recommended solution, and the contact angle of the recommended solution with respect to the fabric 99. The viscosity, surface tension, and contact angle with the fabric 99 of the recommended liquid affect the penetration of the recommended liquid into the fabric 99. Therefore, according to the above configuration, the user can perform appropriate printing on the fabric 99 by selecting a printing liquid based on the physical properties parameters of the recommended liquid presented by the second system 12.

[0051] (3) The acquisition unit 44 acquires pre-print observation data D3 from each of the multiple first systems 11. The output unit 45 outputs data indicating the physical property parameters of the printing solution output by model M1 to the first system 11 from which the pre-print observation data D3 input to model M1 was acquired. With the above configuration, the printing system 10 can present each of the multiple first systems 11 with recommended physical property parameters of the printing solution for the fabric 99.

[0052] (4) The first system 11 includes a display unit 29 that displays the recommended data D6. According to the above configuration, the user can select a printing fluid having the recommended physical properties by checking the display unit 29.

[0053] <Example of changes> The above embodiment can be implemented with the following modifications. The above embodiment and the following modifications can be combined with each other to the extent that they do not contradict each other technically.

[0054] Model M1 may be constructed by having a separate circuit, distinct from the control circuit 42, use training data to train a model through machine learning. The control circuit 42 may perform image analysis on the images included in the observation data D2 and the target image data D5. That is, instead of the control unit 31, the control circuit 42 may extract feature quantities from the fabric 99 or evaluate the image quality.

[0055] The control unit 31 may select a printing fluid based on the physical properties of the printing fluid indicated by the recommended data D6. In this case, the display unit 29 may display information indicating the printing fluid selected by the control unit 31 as the recommended data D6.

[0056] <Technical philosophy> The technical concepts and their effects, as understood from the above-described embodiments and modifications, are explained below.

[0057] (A) The printing system comprises at least one first system including a printing unit that prints on a fabric by discharging a printing liquid onto the fabric and an observation unit that acquires observation data by observing the fabric; and a second system that communicates with the at least one first system via a network, and includes an acquisition unit that acquires pre-printing observation data obtained by the observation unit observing the fabric before printing; a storage unit that stores a model trained by machine learning, which, upon input of the pre-printing observation data acquired by the acquisition unit, outputs physical property parameters of a recommended printing liquid that is recommended for the fabric indicated by the pre-printing observation data; and an output unit that outputs recommended data indicating the physical property parameters output by the model to the at least one first system.

[0058] The pre-print observation data includes information about the fabric. Therefore, according to the above configuration, based on the pre-print observation data, the user can be presented with recommended physical properties of the printing solution for printing on the fabric indicated by the pre-print observation data. This allows the user to select an appropriate printing solution for the fabric.

[0059] (B) In the above printing system, the physical property parameter may include at least one of the viscosity of the recommended liquid, the surface tension of the recommended liquid, and the contact angle of the recommended liquid with respect to the fabric.

[0060] The viscosity, surface tension, and contact angle with the fabric of the recommended liquid affect the penetration of the printing liquid into the fabric. Therefore, with the above configuration, the user can perform appropriate printing on the fabric by selecting a printing liquid based on the physical properties parameters of the recommended liquid presented by the second system.

[0061] (C) In the above printing system, the at least one first system is one of a plurality of first systems, the acquisition unit acquires the pre-print observation data from each of the plurality of first systems, and the output unit may output the recommended data output by the model to the first system among the plurality of first systems that is the source of the pre-print observation data input to the model. With the above configuration, the printing system can present each of the plurality of first systems with recommended physical properties parameters of the printing solution for the fabric.

[0062] (D) In ​​the above printing system, at least one of the first systems may include a display unit for displaying the recommended data. With the above configuration, the user can select a printing solution with recommended physical properties by checking the display unit. [Explanation of Symbols]

[0063] 10...Printing system, 11...First system, 12...Second system, 21...Printing device, 22...Printing section, 23...Head, 24...Nozzle, 25...Carriage, 26...Cartridge, 27...IC chip, 28...Observation section, 29...Display section, 30...Operation section, 31...Control unit, 41...Server device, 42...Control circuit, 43...Storage section, 44...Acquisition section, 45...Output section, 99...Fabric, D1...Printing fluid data, D2...Observation data, D3...Pre-print observation data, D4...Post-print observation data, D5...Target image data, D6...Recommended data, DS...Data set, M1...Model.

Claims

1. A printing unit that prints on the fabric by dispensing a printing solution onto the fabric, An observation unit that acquires observation data by observing the aforementioned fabric, A first system including at least one, A system that communicates with the aforementioned at least one first system via a network, An acquisition unit that acquires pre-printing observation data obtained by the observation unit observing the fabric before printing, A storage unit stores a model trained by machine learning, which, upon input of the pre-print observation data acquired by the acquisition unit, outputs the physical property parameters of a recommended printing solution, which is a recommended printing solution for the fabric indicated by the pre-print observation data. An output unit that outputs recommended data showing the physical property parameters output by the model to at least one first system, The second system is a system that includes, A printing system equipped with the following features.

2. The printing system according to claim 1, characterized in that the physical property parameter includes at least one of the viscosity of the recommended liquid, the surface tension of the recommended liquid, and the contact angle of the recommended liquid with respect to the fabric.

3. The aforementioned at least one first system is one of a plurality of first systems, The acquisition unit acquires the pre-print observation data from each of the plurality of first systems. The printing system according to claim 1, characterized in that the output unit outputs the recommended data output by the model to the first system, which is the source of the pre-print observation data input to the model, among the plurality of first systems.

4. The printing system according to claim 1, characterized in that the at least one first system includes a display unit for displaying the recommended data.

Images