Information processor, information processor control method, and program

Abstract

To determine whether a paper-fed recording medium is an unlearned recording medium.SOLUTION: An information processor has: acquisition means which acquires a characteristic value of a recording medium to which a paper is fed; and determination means which determines that the paper-fed recording medium is an unlearned recording medium when the characteristic value acquired by the acquisition means is not relevant to any recording medium of a plurality of predetermined types of recording media.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present disclosure relates to a technique for estimating the type of a recording medium. [Background technology]

[0002] In the commercial and industrial printing market, the uses of output are diverse, including CAD line drawings, posters, artwork, and signage. For this reason, a wide variety of recording media with different characteristics are used to suit these uses. As the number of types of recording media increases, the task of users selecting the type of recording media fed into a recording device becomes more cumbersome. Some recent recording devices are equipped with a function that improves usability by automatically identifying the type of recording media fed into the device.

[0003] Patent Document 1 discloses a method for determining the type of recording medium from information on the characteristics of a fed recording medium using a trained model based on machine learning. Examples of the information on the characteristics of the recording medium include reflectance, thickness, and basis weight. The trained model is trained using training data that associates information on the characteristics of the recording medium with the type of the recording medium. Therefore, by acquiring information on the characteristics of the fed recording medium and inputting it into the trained model, the type of the fed recording medium can be obtained. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2022-58434 A Summary of the Invention [Problem to be solved by the invention]

[0005] The technology described in Patent Document 1 cannot determine that the fed recording medium is unlearned. With the technology described in Patent Document 1, even if an unlearned recording medium having completely different characteristics from the learned recording medium is fed, the determination result will be one of the learned recording media. Therefore, if there is a discrepancy between the characteristics of the fed recording medium and the characteristics of the recording medium in the determination result, there is a risk of deterioration of usability, printing problems, jams, etc.

[0006] An object of the present disclosure is to determine whether a fed recording medium is an unlearned recording medium. [Means for solving the problem]

[0007] An information processing device according to one aspect of the present disclosure is characterized in having an acquisition means for acquiring characteristic values ​​of a recording medium being fed, and a determination means for determining that the recording medium being fed is an unlearned recording medium if the characteristic value acquired by the acquisition means does not correspond to any of a plurality of predetermined types of recording media. Effect of the Invention

[0008] According to the present disclosure, it is possible to determine whether the fed recording medium is an unlearned recording medium. [Brief description of the drawings]

[0009] [Figure 1] FIG. [Diagram 2] FIG. 2 is a cross-sectional view of a main part of the recording apparatus. [Diagram 3] FIG. 2 is a block diagram showing a control configuration of the printing apparatus. [Figure 4] FIG. 4 is a cross-sectional view of the periphery of the media sensor. [Diagram 5] 10 is a flowchart illustrating an example of a process for estimating the type of a sheet. [Figure 6] FIG. 11 is a diagram showing an example of data obtained by detecting characteristics of a sheet. [Figure 7] FIG. 13 is a diagram illustrating an example of a determination table. [Figure 8] FIG. 13 is a diagram showing an example of an estimation result screen. [Figure 9] 13 is a judgment table for a sheet. [Figure 10] FIG. 2 is a block diagram showing a control configuration of the printing apparatus. [Figure 11] Schematic diagram of a DNN. [Figure 12] 10 is a flowchart illustrating an example of a process for estimating the type of a sheet. [Figure 13] FIG. 13 is a diagram illustrating an estimation result and a similarity. [Figure 14] FIG. 13 is a diagram showing an example of a screen for newly registering an additional recording medium. [Figure 15] 13 is a flowchart showing an example of a process for estimating the type of a sheet S. [Figure 16] FIG. 11 is a diagram showing an example of a selection screen for selecting a type of recording medium. [Figure 17] FIG. 13 is a diagram illustrating an estimation result and a similarity. [Figure 18] 13A and 13B are diagrams illustrating combinations of estimation results of the type of sheet and feature amounts. [Figure 19] 13A and 13B are diagrams illustrating combinations of estimation results of the type of sheet and feature amounts. [Figure 20] FIG. 13 is a diagram showing an example of a screen for accepting a registration method for a recording medium. [Figure 21] FIG. 13 is a diagram illustrating an example of a determination table of characteristic values. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Preferred embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. Note that the following embodiments do not limit the subject matter of the present disclosure, and not all combinations of features described in the following embodiments are necessarily essential to the solutions of the present disclosure. Note that the same components are given the same reference numbers. Furthermore, the relative arrangements, shapes, and the like of the components described in the embodiments are merely examples, and are not intended to limit the scope of the present disclosure to only those.

[0011] In the following description of the embodiment, "recording" includes not only the formation of significant information such as characters and figures, but also the formation of images, designs, patterns, etc. on a sheet. In the present embodiment, a roll sheet is assumed as the sheet, but cut paper, cloth, plastic film, etc. may also be used. Furthermore, "ink" should be broadly interpreted and refers to a liquid that can be applied to a sheet to form an image, design, pattern, etc., or to process the sheet, or to be used for ink processing.

[0012] <<First embodiment>> <Inkjet recording device> FIG. 1 is a perspective view of an inkjet recording device as an example of a recording device 101 that performs industrial / commercial printing in this embodiment. FIG. 2 is a cross-sectional view of the main parts of the recording device 101. FIG. 3 is a block diagram showing the control configuration of the recording device 101. The configuration of the recording device 101 will be described below with reference to FIGS. 1 to 3.

[0013] In this embodiment, as described later, processing using the trained model is performed. In this embodiment, processing using the trained model is performed in the recording device 101. That is, the recording device 101 is an information processing device that uses the trained model. The information processing device that uses the trained model is not limited to the recording device 101. A server (for example, a cloud server) that can exchange various data with the recording device 101 may be used as the information processing device that uses the trained model.

[0014] Hereinafter, the transport direction in which the sheet S is transported in the recording apparatus 101 is defined as the +Y direction. The direction in which the recording head 204 ejects ink onto the sheet S is defined as the -Z direction. The direction in which the recording head 204 moves from the standby position is defined as the +X direction.

[0015] The recording device 101 rotatably holds a roll sheet R in which a sheet S is wound in a roll shape. The roll sheet R is rotated by a roll drive motor 308, and the sheet S is supplied from the roll sheet R to a conveying roller 203. The conveying roller 203 can rotate while holding the sheet S. The conveying roller drive motor 309 rotates the conveying roller 203, and the sheet S is conveyed to a position where the recording head 204 can record on the sheet S. The recording head 204 is mounted on a carriage (not shown) and configured to move back and forth in the X direction. While moving the recording head 204 in the X direction, ink is ejected onto the sheet S conveyed from the recording head 204, thereby recording an image on the sheet S. The sheet S on which the image has been recorded is discharged from a discharge section located downstream of the recording head 204 in the conveying direction, and is loaded onto the basket 103.

[0016] The operation panel 102 is an interface module that accepts various operations from a user. The user can perform various settings for the recording device 101 using various switches or a touch panel provided on the operation panel 102. The various settings for the recording device 101 include, for example, settings for the size and type of the sheet S.

[0017] In the conveying direction, a sheet detection sensor 202 is disposed upstream of the conveying roller 203. When the sheet detection sensor 202 detects that a user has supplied a sheet S from a rolled sheet R, a conveying operation of the sheet S is started. The conveying operation of the sheet S is performed by synchronously driving a roll drive motor 308 and a conveying roller drive motor 309. At this time, the recording apparatus 101 can estimate the type of the sheet S by estimating the sheet type, which will be described later. Details will be described later.

[0018] In the conveying direction, a media sensor 206 and an ultrasonic transmitter 207 are disposed upstream of the sheet detection sensor 202. The media sensor 206 is disposed above the sheet S in the direction of gravity (Z direction), and the ultrasonic transmitter 207 is disposed below the sheet S in the direction of gravity. The media sensor 206 and the ultrasonic transmitter 207 are used to estimate the sheet type, which will be described later.

[0019] An image is recorded on the sheet S as follows. First, the recording device 101 executes a conveying operation to convey the sheet S to a position facing the recording head 204. Next, a recording operation is executed to scan the recording head 204 in a cross direction intersecting (orthogonal to) the conveying direction of the sheet S while discharging ink, thereby recording an image in an area of ​​the sheet S corresponding to the recording head 204. Next, after conveying the sheet S by a predetermined distance, ink is discharged while scanning the recording head 204 in the cross direction. In this way, a desired image is recorded on the sheet S by alternately executing the conveying operation of the sheet S and the image recording operation. The sheet S on which the image has been recorded is sequentially conveyed downstream of the recording head 204 in the conveying direction. The conveyed sheet S is cut by a cutter 205 provided in a discharge section. The cut sheet S is loaded into a basket 103.

[0020] 3 is a block diagram showing an example of the configuration of a control system in the recording apparatus 101. The recording apparatus 101 has an operation panel 102, a recording head 204, a CPU 301, a sensor control unit 302, an input / output interface (IF) 303, a USB port 304, a memory 305, a motor control unit 306, and a RAM 320. The recording apparatus 101 also has a sheet detection sensor 202, a media sensor 206, an ultrasonic transmitter 207, and a carriage encoder 307. The recording apparatus 101 also has a roll drive motor 308, a conveyor roller drive motor 309, a carriage drive motor 310, a lift drive motor 311, a cutter drive motor 312, and a media sensor lift motor 313. The memory 305 has a program 351 and a learned model 352.

[0021] The motor control unit 306 controls each drive motor according to a program 351 recorded in the memory 305. The transport roller drive motor 309 rotates the transport roller 203. The roll drive motor 308 rotates the spool 201. The transport roller drive motor 309 is provided with an encoder that detects the amount of rotation to detect the transport amount of the sheet S. The carriage drive motor 310 rotates a carriage belt (not shown) to move a carriage (not shown) and the recording head 204 mounted on the carriage. The lift drive motor 311 moves the carriage and the recording head 204 up and down. The cutter drive motor 312 drives the cutter. The media sensor lift motor 313 lifts and lowers the media sensor 206.

[0022] Various setting information and the like based on user operations from the operation panel 102 or a host PC connected to a USB port 304 are input to the CPU 301 via an input / output IF 303. The input information is stored in a memory 305. The CPU 301 can read out the information stored in the memory 305 at any time and perform various processes on the read out information. That is, the CPU 301 includes a processing unit that executes various processes. The CPU 301 controls the carriage encoder 307, the sheet detection sensor 202, the media sensor 206, and the ultrasonic transmitter 207 via a sensor control unit 302 to obtain information. The CPU 301 also executes various controls based on inputs from the carriage encoder 307, the sheet detection sensor 202, and the media sensor 206. The RAM 320 is used as a temporary working area.

[0023] <Estimation of seat type> Next, the operation of estimating the type of the sheet S in this embodiment will be described with reference to FIG. 4 to FIG. 7. FIG. 4 is a cross-sectional view of the media sensor 206 and its surroundings. FIG. 5 is a flowchart showing an example of a process for estimating the type of the sheet S. FIG. 6 is a diagram showing an example of data obtained by detecting the characteristics of the sheet S. FIG. 6 is an example of data obtained by detecting the characteristics of the sheet S using the media sensor 206 and the ultrasonic transmitter 207, respectively. FIG. 7 is a judgment table for the sheet S in this embodiment, which is information showing the range of each feature amount corresponding to the type of the sheet S. In this embodiment, the types of the sheet S to be estimated are nine types, namely, recording medium 1 to recording medium 9. That is, the trained model 352 for estimating the type of the sheet S is configured to estimate the recording medium that is most suitable for the input feature amount as the type of the sheet S, among the recording media 1 to recording media 9. The trained model 352 will be described as being stored in the memory 305, but may be provided outside the recording device 101, and the CPU 301 of the recording device 101 may use the trained model provided outside the recording device 101.

[0024] The process of the flowchart shown in FIG. 5 is realized by the CPU 301 of the recording device 101 reading out the program 351 stored in the memory 305 or the like into the RAM 320 and executing it. Note that some or all of the functions of the steps in FIG. 5 may be realized by hardware such as ASIC or electronic circuits. The symbol "S" in the explanation of each process means a step in the flowchart. The process shown in FIG. 5 is started, for example, when the user sets the rolled sheet S in the recording device 101. Alternatively, it may be started when it is detected that the user has set the rolled sheet S in the recording device 101 and then input a predetermined operation by the user on the operation panel 102. The same applies to the flowcharts described below in this specification.

[0025] In S501, the CPU 301 feeds the sheet S. First, the CPU 301 detects that the user has set the rolled sheet R in the recording device 101. Then, the CPU 301 rotates the rolled sheet R by the roll drive motor 308. As a result, the sheet S is supplied from the rolled sheet R to the transport rollers 203. Then, the sheet detection sensor 202 disposed upstream of the transport rollers 203 detects that the sheet S has reached the transport rollers 203. When the sheet detection sensor 202 detects that the sheet S has reached the transport rollers 203, the CPU 301 stops driving the roll drive motor 308. At the position where the sheet S is detected by the sheet detection sensor 202, the sheet S is transported to a position facing the media sensor 206 and the ultrasonic transmitter 207.

[0026] In S502, the CPU 301 performs sensing. That is, the CPU 301 measures the characteristics of the sheet S by controlling the media sensor 206 and the ultrasonic transmitter 207 via the sensor control unit 302. As shown in FIG. 4, the media sensor 206 has a CIS (Contact Image Sensor) 401 and a microphone 402. A roller 403 is disposed at a position facing the CIS 401. The ultrasonic transmitter 207 is disposed at a position facing the microphone 402. The CPU 301 lowers the media sensor 206, which is separated from the sheet S, by the media sensor lift motor 313, so that the sheet S can be sandwiched between the CIS 401 and the roller 403. By sandwiching the sheet S, the characteristics of the sheet S can be stably measured. The CPU 301 reads the surface image of the sheet S by the CIS 401 while conveying the sheet S again in a state where the sheet S is sandwiched between the CIS 401 and the roller 403. Then, when the sensing is completed, the CPU 301 separates the media sensor 206 from the sheet S by raising the media sensor 206 using the media sensor lift motor 313.

[0027] 6(a) and (b) show examples of surface images of sheet S acquired using the CIS 401. Fig. 6(c) and (d) show examples of electrical signals of ultrasonic waves transmitted through sheet S acquired using the ultrasonic transmitter 207 and microphone 402.

[0028] The CIS 401 is a line sensor extending in the width direction of the sheet S, and acquires one-dimensional (one line) image data. The CPU 301 acquires image data of the sheet S using the CIS 401 while synchronously driving the roll drive motor 308 and the conveying roller drive motor 309 in a state where the sheet S is sandwiched between the CIS 401 and the roller 403. By reading the image of the sheet S using the CIS 401 while conveying the sheet S in this way, two-dimensional image data as shown in Figs. 6(a) and (b) can be acquired. Fig. 6(a) is an example of a surface image of Japanese paper, and Fig. 6(b) is an example of a surface image of synthetic paper. In Figs. 6(a) and (b), the CIS direction corresponds to the width of the CIS 401 (width in the X direction intersecting with the sheet S), and the conveying direction corresponds to the conveying amount of the sheet S measured by the CIS 401. Note that, although an example of measurement using a one-dimensional sensor as the CIS 401 is shown here, the surface image of the sheet S may be measured using a two-dimensional sensor. Moreover, when the CIS 401 is measuring the surface image, the ultrasonic transmitter 207 and the microphone 402 (sound pickup sensor) obtain ultrasonic electrical signals as shown in Figs. 6(c) and (d). Fig. 6(c) is an example of an ultrasonic electrical signal transmitted through Japanese paper, and Fig. 6(d) is an example of an ultrasonic electrical signal transmitted through synthetic paper. In this embodiment, an example of obtaining an ultrasonic electrical signal in conjunction with the measurement of the surface image by the CIS 401 will be described, but this is not limiting. The measurement of the surface image and the acquisition of the ultrasonic electrical signal may be performed separately. Moreover, the ultrasonic electrical signal may be obtained in a state in which the sheet S is not being transported.

[0029] In S503, CPU 301 derives feature amounts related to surface information and feature amounts related to cross-sectional information of sheet S from the characteristics of sheet S measured in S502 using a feature amount derivation method previously stored in memory 305. That is, CPU 301 derives feature amounts related to surface information and feature amounts related to cross-sectional information of sheet S from the surface image of sheet S and the electrical signal of the ultrasonic waves.

[0030] The CPU 301 derives three feature amounts relating to the surface information of the sheet S from the surface image of the sheet S as shown in FIGS. 6(a) and (b). The first feature amount is brightness. The brightness is derived as the average value of all pixel values ​​in the surface image. The second feature amount is unevenness in the CIS direction. The unevenness in the CIS direction is derived as the average value of the absolute values ​​of the differences between adjacent pixel values ​​in the CIS direction in the surface image. The third feature amount is unevenness in the transport direction. The unevenness in the transport direction is derived as the average value of the absolute values ​​of the differences between adjacent pixel values ​​in the transport direction in the surface image.

[0031] The CPU 301 derives three feature amounts relating to the cross-sectional information of the sheet S from the electrical signal of the ultrasonic waves transmitted through the sheet S as shown in FIGS. 6(c) and (d). The first feature amount is peak 1. Peak 1 is derived as the maximum voltage value in the period from time t1 to time t2. The second feature amount is peak 2. Peak 2 is derived as the maximum voltage value in the period from time t2 to time t3. The third feature amount is peak 3. Peak 3 is derived as the maximum voltage value in the period from time t3 to time t4. Note that although the maximum voltage value is described here, it may also be the minimum voltage value. The cross-sectional information of the sheet S corresponds to information such as the thickness and basis weight of the sheet.

[0032] In this manner, in S503, the CPU 301 can derive six feature amounts relating to the surface information and cross-sectional information of the sheet S from the characteristics of the sheet S measured in S502.

[0033] In this embodiment, Japanese paper and synthetic paper are taken as examples. An example of the relationship between the feature quantities of the two is shown below. The luminance is higher for synthetic paper than for Japanese paper. The unevenness in the CIS direction is higher for Japanese paper than for synthetic paper. The unevenness in the transport direction is higher for Japanese paper than for synthetic paper. For example, a sheet S with a whiter color and flatter surface has a higher luminance. Also, the greater the unevenness of the surface, the greater the unevenness in both the CIS direction and the transport direction. Note that some types of sheets have longitudinal or transverse grains, and only one type of unevenness is greater. The thicker the sheet is, the smaller the peak value of the ultrasonic electrical signal becomes. Japanese paper is thicker than synthetic paper. For this reason, the peak value is larger for synthetic paper than for Japanese paper. Also, even with the same thickness, the peak value changes depending on the cross section (the material or density constituting the sheet). Specifically, the peak value tends to be smaller when a material (medium) with a larger acoustic impedance is used.

[0034] Next, in S504, the CPU 301 estimates the type of the sheet S from the six feature amounts derived in S503 using the trained model 352 stored in advance in the memory 305. The trained model 352 is trained using training data that is a set of the six feature amounts related to the characteristics of the sheet S derived in S503 and the type of the sheet S associated with them. In this embodiment, as described above, it is assumed that training is performed in the trained model 352 using the training data corresponding to the recording medium 1 to the recording medium 9 shown in FIG. 7. Therefore, by inputting the six feature amounts related to the characteristics of the sheet S to the trained model 352, the type of the sheet S can be obtained as an output from the trained model 352.

[0035] In this way, in S504, CPU 301 can estimate the type of sheet S by using the six feature amounts derived in S503. However, the estimation result using trained model 352 will be any one of recording media 1 to 9, which are the types of sheet S to be estimated. Therefore, even if an unlearned sheet S having completely different characteristics from recording media 1 to 9, which are the trained types of sheet S, is fed, the estimation result will be any one of recording media 1 to 9, and it cannot be determined that it is unlearned. Hereinafter, recording media 1 to 9 will be referred to as trained sheet types. Moreover, sheet types other than recording media 1 to 9 will be referred to as unlearned sheet types.

[0036] In S505, the CPU 301 uses the judgment table shown in FIG. 7, which is stored in advance in the memory 305, to judge whether the fed sheet S to be processed is an unlearned sheet type or not, based on the six feature amounts derived in S503.

[0037] FIG. 7 is a diagram showing an example of the judgment table in this embodiment. In this embodiment, an example in which the judgment table is stored in the memory 305 will be described, but the judgment table may be stored outside the recording device 101. The judgment table shown in FIG. 7 includes minimum and maximum values ​​as ranges of each feature amount corresponding to the learned sheet types, recording medium 1 to recording medium 9. These data are obtained by repeatedly measuring the characteristics of recording medium 1 to recording medium 9, deriving six feature amounts from the measurement results, and deriving the minimum and maximum values ​​of each. That is, the judgment table shown in FIG. 7 stores data obtained by measuring the recording medium in advance.

[0038] 7, it is possible to specify the range in which each feature amount of the sheet S derived in S503 should fall with respect to the type of the sheet S estimated in S504 described above. If all feature amounts derived in S503 fall within the ranges of the judgment table, CPU 301 determines that the fed sheet S to be processed is of a learned sheet type. If one or more feature amounts do not fall within the ranges of the judgment table, CPU 301 determines that the fed sheet S to be processed is not of a learned sheet type. In other words, CPU 301 determines that the fed sheet S to be processed is of an unlearned sheet type.

[0039] As an example, a case will be described in which the estimation result of the sheet S in S504 is the recording medium 1. It is assumed that the feature amount of the sheet S derived in S503 is luminance 1.7, unevenness in the CIS direction 0.025, unevenness in the transport direction 0.02, peak 1 0.13, peak 2 0.25, and peak 3 0.33. In the case of such feature amounts, all feature amounts are within the range corresponding to the recording medium 1 in the judgment table of FIG. 7. Therefore, in this case, the CPU 301 can determine that the sheet S to be processed that is being fed is a learned sheet type. On the other hand, for example, if the luminance is 1.8, one or more feature amounts are not within the range of the judgment table of FIG. 7. Therefore, in this case, the CPU 301 can determine that the sheet S to be processed that is being fed is an unlearned sheet type. Similarly, when other feature amounts are not within the range of the judgment table, the CPU 301 can determine that the sheet S to be processed that is being fed is an unlearned sheet type.

[0040] As described above, in S505, the CPU 301 can determine whether or not the fed sheet S to be processed is an unlearned sheet type, using the six feature amounts derived in S503 and the determination table in FIG.

[0041] Next, in S506, the CPU 301 displays the estimation result of the sheet S on the operation panel 102 via the input / output IF 303, thereby presenting the estimation result of the sheet S to the user. The estimation result of the sheet S is a combination of the estimation result of S504 and the judgment result of S505 described above. The estimation result of S504 is one of the recording media 1 to 9 that is the most appropriate type of the sheet S. The judgment result of S505 is whether the sheet S is an unlearned sheet type. The estimation result of the sheet S is presented to the user via the operation panel 102. If the sheet S is an unlearned sheet type, a warning is given to the user via the operation panel 102. As a warning, for example, the user may be prompted to review the type of the sheet S and the printing parameters set in the recording device 101. Alternatively, the user may be prompted to add the fed sheet S to the types of the sheet S to be estimated so that the fed sheet S becomes a learned sheet type.

[0042] FIG. 8 is a diagram showing an example of an estimation result screen that the CPU 301 displays on the operation panel 102 in S506. The estimation result screen 800 shown in FIG. 8(a) includes a message 801 that indicates to the user that a recording medium called "my PVC 1" is estimated as the estimation result of S504 and that "my PVC 1" is being fed. Also, in FIG. 8(a), the determination result of S505 is presented as a learned sheet type. In the example of FIG. 8, a warning icon is displayed when the sheet is an unlearned sheet. Therefore, based on the fact that the warning icon is not displayed, the determination result of S505 is presented as a learned sheet type. The estimation result screen 850 shown in FIG. 8(b) includes a message 801 that indicates to the user that "my PVC 1" is estimated as the estimation result of S504 and that "my PVC 1" is being fed. 8(b) includes a warning icon 851 indicating that the sheet is an unlearned sheet as a result of the determination in S505. When the warning icon 851 is pressed by the user, as described above, the screen may transition to a screen that notifies the user of a need to review the type of sheet S and printing parameters, or such a message may be displayed on a pop-up screen.

[0043] As described above, according to this embodiment, it is possible to determine whether the fed recording medium is an unlearned recording medium. In this embodiment, not only is the type of the sheet S estimated, but also each characteristic amount of the sheet S is compared with each characteristic amount of the learned sheet type, thereby making it possible to determine that the sheet S is an unlearned sheet type.

[0044] <<Second embodiment>> In the first embodiment, an example was described in which, in order to determine whether the sheet S is an unlearned sheet type, it is determined whether each feature falls within the range of the determination table using the determination table shown in FIG. 7. In this embodiment, an example is described in which it is determined whether the sheet is an unlearned sheet type using the distance between data when two feature amounts are combined. Specifically, an example is described in which it is determined whether the sheet is an unlearned sheet type using the Mahalanobis distance when two feature amounts are combined. The difference from the first embodiment is the process of S505 in the flowchart shown in FIG. 5. Other configurations and the like can be the same as the example described in the first embodiment. Hereinafter, in this embodiment, the differences will be mainly described.

[0045] <Estimation of seat type> The operation of estimating the type of sheet S in this embodiment will be described with reference to Fig. 5 and Fig. 9. Fig. 5 is a flowchart described in the first embodiment. Fig. 9 is a judgment table for sheet S in this embodiment, showing the range of Mahalanobis distances in combinations of feature amounts corresponding to the type of sheet S. In this embodiment, as in the first embodiment, the types of sheet S to be estimated are nine types, from recording medium 1 to recording medium 9.

[0046] In the process of FIG. 5, steps S501 to S504 and S506 are the same as those described in the first embodiment, and therefore a description thereof will be omitted here.

[0047] In S505, the CPU 301 uses the judgment table shown in FIG. 9 stored in advance in the memory 305 to judge whether the sheet S is an unlearned sheet type from the six feature amounts derived in S503. FIG. 9 is an example of the judgment table in this embodiment. The judgment table shown in FIG. 99 stores the range of the Mahalanobis distance in the combination of each feature amount corresponding to the recording medium 1 to the recording medium 9. These data are obtained by repeatedly measuring the characteristics of the recording medium 1 to the recording medium 9, deriving the six feature amounts from the measurement results, and calculating the maximum Mahalanobis distance for each combination of the feature amounts. In addition to the judgment table in FIG. 9, the average value of each feature amount required to calculate the Mahalanobis distance and the variance-covariance matrix of the combination of each feature amount are stored in the memory 305. The method of calculating the Mahalanobis distance is publicly known, so a description thereof will be omitted here.

[0048] 9, it is possible to specify within what range the Mahalanobis distance calculated by combining the features of the sheet S derived in S503 should fall for the type of the sheet S estimated in S504 described above. If all feature combinations fall within the ranges of the judgment table in FIG. 9, the CPU 301 determines that the fed sheet S to be processed is of a learned sheet type. If one or more feature combinations do not fall within the ranges of the judgment table in FIG. 9, the CPU 301 determines that the fed sheet S to be processed is not of a learned sheet type. In other words, the CPU 301 determines that the fed sheet S to be processed is of an unlearned sheet type.

[0049] As an example, a case where the estimation result of the sheet S in S504 is the recording medium 1 will be described. Assume that the Mahalanobis distance calculated from the feature amount of the sheet S derived in S503 and the average value of each feature amount and the variance-covariance matrix of the combination of each feature amount stored in the memory 305 is the following value. That is, assume that the Mahalanobis distance is 3.8 for brightness vs. peak 1, 3.1 for unevenness in the CIS direction vs. peak 2, and 3.6 for unevenness in the conveying direction vs. peak 3. In this case, all the combinations of the feature amounts fall within the range corresponding to the recording medium 1 in the determination table of FIG. 9. Therefore, the CPU 301 can determine that the fed sheet S to be processed is a learned sheet type. On the other hand, for example, if the brightness vs. peak 1 is 4.5, one or more combinations of the feature amounts do not fall within the range of the determination table of FIG. 9. Therefore, in this case, the CPU 301 can determine that the fed sheet S to be processed is an unlearned sheet type. Similarly, when the combination of other feature amounts is not within the range of the determination table, the CPU 301 can determine that the fed sheet S to be processed is of an unlearned sheet type.

[0050] The judgment table in FIG. 9 is merely an example, and is not limited to this example. The number of combinations of feature amounts is not limited to the three combinations shown in FIG. 9, and may be more than three or less than three. The combinations themselves are not limited to the example shown in FIG. 9. For example, a combination of feature amounts of surface information such as "brightness vs. unevenness in the CIS direction" or a combination of feature amounts of cross-sectional information such as "peak 1 vs. peak 2" may be used. In addition, in the judgment table of this embodiment, an example in which three types of sets of two feature amounts are prepared has been described, but a plurality of types of sets of three or more feature amounts may be prepared.

[0051] In this manner, in S505 in this embodiment, the CPU 301 can use the six feature amounts derived in S503 and the judgment table of FIG. 9 to determine whether the sheet S being fed to be processed is an unlearned sheet type.

[0052] As described above, in this embodiment, the type of sheet S is not only estimated, but also it is possible to determine that the sheet S is an unlearned sheet type by comparing a combination of each feature amount of the sheet S with a combination of each feature amount of a learned sheet type. In this embodiment, the determination is made using a combination of each feature amount, so that it is possible to improve the accuracy of determining that the sheet type is an unlearned sheet type compared to the first embodiment in which the determination is made using each feature amount alone. In addition, in this embodiment, the Mahalanobis distance is used instead of the Euclidean distance that is generally used as a distance. Therefore, the distance of the combination of each feature amount is obtained taking into account the correlation of data, so that it is possible to further improve the accuracy of the determination.

[0053] <<Third embodiment>> In the first and second embodiments, in order to determine whether the sheet S is an unlearned sheet type, an example was described in which the determination table shown in FIG. 7 or FIG. 9 is used to determine whether each feature amount or a combination of each feature amount falls within the range of the determination table. In this embodiment, a machine learning trained model is used to determine whether the sheet S is an unlearned sheet type. The difference from the first embodiment is the process of S505 in the flowchart shown in FIG. 5. Also, in addition to the trained model 352 described in the first embodiment, the second trained model is used. Other configurations and the like can be the same as the example described in the first embodiment. Hereinafter, in this embodiment, the differences will be mainly described.

[0054] Fig. 10 is a block diagram showing the control configuration of the recording device 101 in this embodiment. In Fig. 10, in addition to the trained model 352, the memory 305 includes a second trained model 1001 as a separate trained model. The other basic configurations are similar to those of the example described in the first embodiment.

[0055] <Estimation of seat type> The operation of estimating the type of sheet S in this embodiment will be described with reference to Fig. 5. Fig. 5 is a flowchart described in the first embodiment. In this embodiment, similar to the first embodiment, the types of sheet S to be estimated are nine types, from recording medium 1 to recording medium 9.

[0056] In the process of FIG. 5, steps S501 to S504 and S506 are the same as those described in the first embodiment, and therefore a description thereof will be omitted here.

[0057] In S505, the CPU 301 uses the second trained model 1001 stored in advance in the memory 305 to determine whether the fed sheet S to be processed is an unlearned sheet type from the six feature amounts derived in S503. In S505 of this embodiment, three combinations of two feature amounts are prepared, and a dedicated trained model (the second trained model 1001) corresponding to each combination is used to determine whether the sheet S is an unlearned sheet type or a trained sheet type. That is, in this embodiment, three trained models are prepared as the second trained model 1001. The first combination is luminance and peak 1. The values ​​of luminance and peak 1 are input to the 2-1 trained model, and it is determined whether the sheet S is an unlearned sheet type or a trained sheet type. The second combination is unevenness in the CIS direction and peak 2. The values ​​of unevenness in the CIS direction and peak 2 are input to the 2-2 trained model, and it is determined whether the sheet S is an unlearned sheet type or a trained sheet type. The third combination is unevenness in the transport direction and peak 3. The values ​​of unevenness in the transport direction and peak 3 are input into the 2-3 learned model, and it is determined whether the sheet S is an unlearned sheet type or a learned sheet type. If the determination results of all feature combinations are a learned sheet type, the CPU 301 determines that the fed sheet S to be processed is a learned sheet type. If the determination result of one or more feature combinations is an unlearned sheet type, the CPU 301 determines that the fed sheet S to be processed is an unlearned sheet type.

[0058] For example, assume that the determination result based on the luminance and peak 1 is a learned sheet type, the determination result based on the unevenness in the CIS direction and peak 2 is a learned sheet type, and the determination result based on the unevenness in the transport direction and peak 3 is a learned sheet type. In this case, the determination results of all feature amount combinations are learned sheet types, so the CPU 301 can determine that the sheet S is a learned sheet. Also, if the determination result of the sheet S based on the luminance and peak 1 is an unlearned sheet type, the determination results of one or more feature amount combinations are unlearned sheet types, so the CPU 301 can determine that the sheet S is an unlearned sheet type. Similarly, if the determination results based on other feature amount combinations are unlearned sheet types, the CPU 301 can determine that the sheet S is an unlearned sheet type.

[0059] In this manner, in S505 of the present embodiment, the CPU 301 can use the six feature amounts derived in S503 and the second trained model 1001 to determine whether or not the sheet S is an untrained sheet type.

[0060] In this embodiment, an example in which the number of the second trained models 1001 is three has been described, but this is not limited thereto. The number may be more than three or less than three. The combination itself is also not limited to the above-mentioned example. For example, a combination of features of surface information such as "brightness vs. unevenness in the CIS direction" or a combination of features of cross-sectional information such as "peak 1 vs. peak 2" may be used. In this embodiment, an example in which the input data to the second trained model 1001 is two features has been described, but three or more features may be used as input data to the second trained model 1001.

[0061] As described above, in this embodiment, not only is the type of sheet S estimated, but by using the judgment results of each learned model for the combination of each feature of sheet S, it is possible to determine that sheet S is an unlearned sheet type.

[0062] <Pre-trained model> 11 is a schematic diagram of a deep neural network (DNN). Three second trained models 1001 used in the above-described S505 of this embodiment will be described with reference to FIG.

[0063] The second trained model 1001 of this embodiment is a DNN as shown in the schematic diagram of FIG. 11. The DNN receives data at an input layer 1101, propagates the data through an intermediate layer 1102, and outputs the data at an output layer 1103. Each layer has a plurality of nodes indicated by circles. The input data is propagated toward the output layer while being weighted, biased, and the like between the nodes of each layer. Adjusting parameters such as weighting and bias so that a specified output is obtained for a specified input is expressed as learning a model. A trained model is called a trained model. A data set of input data and output data associated with the input data used to train a model is called teacher data.

[0064] The input data of the teacher data in this embodiment is a combination of two feature amounts out of six feature amounts derived from the characteristics of the sheet S. The six feature amounts are the same as those derived in the feature amount derivation process of S503, and are brightness, unevenness in the CIS direction, unevenness in the transport direction, peak 1, peak 2, and peak 3. In this embodiment, a different combination of feature amounts is used for each model to be trained. In addition, the input layer 1101 of the trained model has two nodes. Each feature amount is input to each node.

[0065] In this embodiment, the output data of the teacher data is an integer value indicating whether the sheet S is an unlearned sheet type or not, and is 0 and 1. When actually learning the model, the integer value converted into a one-hot vector is used. In addition, the output layer 1103 of the learned model has two nodes. The nodes of the output layer 1103 output the probability that the sheet S is an unlearned sheet type and the probability that the sheet S is a learned sheet type, respectively. If the output of the learned model is considered as an array, each element of the output array can be considered to be the above two probabilities. If each element of the output array is linked to the above two probabilities, the index of each element is also linked to the above two probabilities. In this embodiment, the estimation result of the learned model is the index of the element with the highest probability. In this embodiment, a learned sheet type is indicated by the integer value 0, and an unlearned sheet type is indicated by the integer value 1.

[0066] The first second trained model 1001 is a 2-1 trained model that determines whether or not the sheet S is an unlearned sheet type from the luminance and peak 1, which are the feature quantities of the sheet S. The input data of the teacher data of the 2-1 trained model is the luminance and peak 1. As input data for learning the learned sheet type, the characteristics of the recording medium 1 to the recording medium 9 are repeatedly measured, and the luminance and peak 1 are derived from each measurement data. In addition, as input data for learning the unlearned sheet type, dummy data of the luminance and peak 1 that imitates the unlearned sheet type is generated. The output data corresponding to the input data for learning the learned sheet type is an integer value of 0, indicating that it is a learned sheet type. The output data corresponding to the input data for learning the unlearned sheet type is an integer value of 1, indicating that it is an unlearned sheet type.

[0067] The model trained using the above training data is referred to as the 2-1 trained model. The 2-1 trained model can determine whether or not the sheet S is an unlearned sheet type from the brightness and peak 1, which are the feature quantities of the sheet S. If the sheet S is a learned sheet type, 0 is output as the estimated result. If the sheet S is an unlearned sheet type, 1 is output as the estimated result.

[0068] The second trained model 1001 is a trained model 2-2 that determines whether the sheet S is an unlearned sheet type from the unevenness in the CIS direction and peak 2, which are the features of the sheet S. The input data of the teacher data of the trained model 2-2 is the unevenness in the CIS direction and peak 2. As input data for learning the trained sheet type, the characteristics of the recording medium 1 to the recording medium 9 are repeatedly measured, and the unevenness in the CIS direction and peak 2 are derived from each measurement data. In addition, as input data for learning the unlearned sheet type, dummy data of the unevenness in the CIS direction and peak 2 that imitates the unlearned sheet type is generated. The output data corresponding to the input data for learning the trained sheet type is an integer value of 0, indicating that it is a trained sheet type. The output data corresponding to the input data for learning the unlearned sheet type is an integer value of 1, indicating that it is an unlearned sheet type.

[0069] The model trained using the above training data is referred to as the 2-2 trained model. The 2-2 trained model can determine whether or not the sheet S is an unlearned sheet type from the unevenness in the CIS direction and peak 2, which are the features of the sheet S. If the sheet S is a trained sheet type, 0 is output as the estimation result. If the sheet S is an unlearned sheet type, 1 is output as the estimation result.

[0070] The third trained model is a 2-3 trained model that determines whether or not the sheet S is an unlearned sheet type from the unevenness in the transport direction and peak 3, which are the feature quantities of the sheet S. The input data of the teacher data of the 2-3 trained model are the unevenness in the transport direction and peak 3. As input data for learning the learned sheet type, the characteristics of the recording medium 1 to the recording medium 9 are repeatedly measured, and the unevenness in the transport direction and peak 3 are derived from each measurement data. In addition, as input data for learning the unlearned sheet type, dummy data of the unevenness in the transport direction and peak 3 that imitates the unlearned sheet type is generated. The output data corresponding to the input data for learning the learned sheet type is an integer value of 0, indicating that the sheet type is a learned sheet type. The output data corresponding to the input data for learning the unlearned sheet type is an integer value of 1, indicating that the sheet type is an unlearned sheet type.

[0071] The model trained using the above teacher data is referred to as the 2-3 trained model. The 2-3 trained model can determine whether or not the sheet S is an unlearned sheet type from the unevenness in the transport direction and peak 3, which are the feature quantities of the sheet S. If the sheet S is a learned sheet type, 0 is output as the estimation result. If the sheet S is an unlearned sheet type, 1 is output as the estimation result.

[0072] The above three second trained models 1001 are generated, for example, by a machine learning device (not shown). The machine learning device is, for example, a PC. The machine learning device can record the trained models and a calculation method required for estimation using the trained models in the memory 305 via the USB port 304, the input / output IF 303, and the CPU 301. The CPU 301 measures the characteristics of the sheet S using the media sensor 206 and the ultrasonic transmitter 207, and derives features related to the surface information and cross-sectional information of the sheet S from the measurement data. Then, the CPU 301 can input the derived features to the trained model to determine whether the sheet S is an untrained sheet type.

[0073] <<Fourth embodiment>> In the first to third embodiments, an example of determining whether the sheet S is an unlearned sheet type has been described. In this embodiment, a similarity is derived from a characteristic value detected by sensing, and the type of the sheet S is determined based on a combination of the estimation result of the sheet type and the similarity. Furthermore, the combination of the estimation result and the similarity is stored as the type of the added recording medium. As a result, it is possible to determine the type of the sheet S, including the type of the added recording medium. That is, in this embodiment, when there is an unlearned sheet type, a new sheet type can be added. In addition, when this addition is made, the existing learned model can be continued to be used without reconstructing (re-learning) the learned model used in the estimation process. Since the basic configuration is the same as the example described in the first embodiment, the description here is omitted, and the differences are mainly described.

[0074] <Estimation of seat type> Fig. 12 is a flowchart showing an example of a process for estimating the type of sheet S. Fig. 13 is a diagram for explaining the estimation result and the similarity in this embodiment. In Fig. 12, steps S1201 to S1204 are the same as steps S501 to S504, and therefore will not be described here.

[0075] In S1205, the CPU 301 derives a similarity from the characteristic value derived (detected) in S1203. As an example, the CPU 301 determines whether the characteristic value detected in S1203 is included in a predetermined range stored in the judgment table shown in FIG. 7 for the recording medium estimated in S1204. Then, the CPU 301 derives a numerical value representing each characteristic as 0 and 1 as the similarity, with 0 being the case where the characteristic value is included in the predetermined range and 1 being the case where the characteristic value is not included. For example, FIG. 13(a) is an example of a result of determining whether the recording medium estimated in S1204 is recording medium 4, and whether the characteristic value detected in S1203 is included in the range stored for recording medium 4 in the judgment table shown in FIG. 7. For example, when the unevenness in the CIS direction and peak 2 are not included in the predetermined range, the similarity is derived as shown in FIG. 13(a).

[0076] In this embodiment, in addition to the judgment table shown in Fig. 7, a combination table of the estimation result and the similarity as shown in Fig. 13(b) is stored in, for example, the memory 305. Fig. 13(b) is a table in which all items are stored with a similarity of 0 for all recording media. In other words, Fig. 13(b) is a diagram showing contents substantially equivalent to those of the judgment table shown in Fig. 7. In this embodiment, the type of recording medium is determined based on the combination of the estimation result and the similarity. To be used for this determination, a combination table of the estimation result and the similarity as shown in Fig. 13(b) is prepared.

[0077] In S1206, CPU 301 determines whether there is a type of recording medium that matches the combination of the estimation result obtained in S1204 and the similarity derived in S1205 in a combination table such as that shown in Fig. 13(b). If a type of recording medium whose similarity matches the estimation result exists in the combination table, the matching type of recording medium is determined as the estimation result. On the other hand, if a type of recording medium whose similarity matches the estimation result does not exist in the combination table, in the example described in the first embodiment, it was merely determined to be an unlearned sheet, but in this embodiment, it is also possible to add it as a new type of recording medium.

[0078] In S1207, the CPU 301 displays the estimation result of the sheet S on the operation panel 102 via the input / output IF 303, thereby presenting the estimation result of the sheet S to the user. As described above, the estimation result of the sheet S is determined based on the combination of the estimation result of S1204 and the similarity of S1205, and the combination table. The estimation result of S1204 is one of the recording media 1 to 9 that is the most appropriate type of the sheet S. In the determination result of S1206, if there is a matching type of recording medium in the combination table of the estimation result and similarity, the sheet S is presented to the user as a learned sheet type, and a message 801 is displayed in the estimation result screen 800 as shown in FIG. 8(a). On the other hand, if there is no matching recording medium in the combination table of the estimation result and similarity, the estimation result is displayed as shown in the message 801 as shown in FIG. 8(b), and a warning icon 851 indicating that the sheet type is unlearned is displayed. FIG. 8 is the same as the example described in the first embodiment.

[0079] Next, in S1208, CPU 301 determines whether to register the recording medium. Specifically, when a warning icon 851 indicating an unlearned sheet type is pressed by the user on estimation result screen 850 shown in Fig. 8(b), CPU 301 determines to register the recording medium. On the other hand, when warning icon 851 is not pressed and another icon or the like is pressed, CPU 301 determines not to register the recording medium.

[0080] FIG. 14 is a diagram showing an example of a screen for newly registering a recording medium. When the warning icon 851 is pressed, the CPU 301 displays a name input screen 1400 as shown in FIG. 14. When the CPU 301 accepts the input of a name by the user, the CPU 301 stores the additional storage medium in the combination table shown in FIG. 13(b). FIG. 13(c) is a combination table in a state in which the recording medium 10 is newly added to FIG. 13(b). The name of the recording medium accepted in the name input screen 1400 is stored in the recording medium 10. In addition, in the record of the recording medium 10, the recording medium 4 estimated in S1204 is stored in the estimation result, and the similarity derived in S1205 is stored in the similarity. Then, in the next and subsequent flowcharts, processing using the combination table shown in FIG. 13(c) is performed. Therefore, in the next estimation process, if the estimation result and similarity corresponding to the recording medium 10 shown in FIG. 13(c) are derived, the type of the recording medium is determined to be the recording medium 10.

[0081] In this embodiment, an example has been described in which a new type of recording medium is registered in response to pressing of the warning icon 851, but the present invention is not limited to this example. Even in a state in which the warning icon 851 is not displayed, paper may be added and registered from a setting screen (not shown).

[0082] As described above, in this embodiment, the type of sheet S is determined from a combination table of the sheet estimation result and the similarity. This allows even an unlearned sheet type to be stored as an additional recording medium, and in the subsequent estimation process, it becomes possible to determine the type of the additional recording medium as well. Furthermore, even if a recording medium is added, the type of the recording medium can be estimated without reconstructing (re-learning) the existing trained model.

[0083] <<Fifth embodiment>> In the fourth embodiment, an example was described in which the similarity is derived from the characteristic value of the sheet S, and the type of the sheet is determined from a combination table of the estimation result and similarity of the sheet S. Also, in the fourth embodiment, an example was described in which the estimation result and similarity are stored as an additional sheet, so that the sheet type can be identified even for an unlearned sheet type. On the other hand, there may be cases in which the sheet is not determined to be the original sheet type in the determination using the estimation result and similarity due to sheet deterioration, lot difference, or environmental change. In such cases, it is preferable to determine the sheet type as one of the existing sheet types rather than adding a new sheet type. In this embodiment, the estimation result and similarity are used as in the example described in the fourth embodiment. Then, in the combination table of the estimation result and similarity, the similarity linked to the recording medium of the estimation result is stored in a new area in the combination table based on a user instruction. Then, an example will be described in which the newly stored area is also referenced during the determination, so that it is possible to determine that the recording medium is the same even if the recording medium is deteriorated, the lot difference, or the environmental change occurs.

[0084] Since the basic configuration is similar to the example described in the fourth embodiment, the description will be omitted here and the differences will be mainly described.

[0085] <Estimation of seat type> Fig. 15 is a flow chart showing an example of a process for estimating the type of sheet S. Fig. 16 is a diagram showing an example of a selection screen for selecting the type of recording medium. Fig. 17 is a diagram for explaining the estimation result and the similarity in this embodiment. In Fig. 15, steps S1501 to S15005 are similar to steps S1201 to S1205, and therefore description thereof will be omitted here.

[0086] In S1506, the CPU 301 determines whether there is a recording medium that matches the combination of the estimation result obtained in S1504 and the similarity derived in S1505 in a combination table such as that shown in Fig. 17. The table in Fig. 17 will be described later.

[0087] In S1507, the CPU 301 displays the estimation result of the sheet S on the operation panel 102 via the input / output IF 303, thereby presenting the estimation result of the sheet S to the user.

[0088] Next, in S1508, CPU 301 determines whether to link the estimation result and the similarity to a recording medium. Specifically, when a warning icon 851 indicating an unlearned sheet type is pressed by the user on estimation result screen 850 shown in Fig. 8(b), CPU 301 determines to link to a recording medium. On the other hand, when warning icon 851 is not pressed and another icon is pressed, CPU 301 determines not to link to a recording medium.

[0089] FIG. 16 is a diagram showing an example of a selection screen for selecting the type of recording medium. When the warning icon 851 is pressed, the CPU 301 displays a selection screen 1600 as shown in FIG. 14. The selection screen 1600 in FIG. 14 is a screen configured to be able to accept the selection of any paper type from among existing paper types from the user. When the paper type is selected by the user, the combination of the estimation result and the similarity is linked and stored in a new storage area in the selected corresponding recording medium. The combination table in FIG. 17 shows an example in which the association with the recording medium is performed from area 0 to area 9. In this example, area 0 is a table in which all items are stored as similarity 0 in all recording media. In other words, area 0 in FIG. 17 is a diagram showing contents substantially equivalent to the judgment table shown in FIG. 7. Area 1 in FIG. 17 is an area in which the estimation result is recording medium 4, and the result that the CIS direction response and peak 2 are not similar in similarity is linked, as in the example described in the fourth embodiment. That is, although it was determined to be an unlearned sheet, the user selected it as a learned sheet (i.e., as recording medium 4), and as a result, the values ​​are stored as shown in area 1 in Fig. 17. That is, although the values ​​of the CIS direction responsiveness and peak 2 are out of range, the user selected it as a known type of recording medium under the judgment that this is due to deterioration of the recording medium, lot differences, or environmental changes. In this way, the combination table in Fig. 17 is in a state in which items are stored up to area 9.

[0090] In this embodiment, an example has been described in which a selection screen is displayed when the warning icon 851 indicating an unlearned sheet type is pressed by the user, but this is not limiting. Even if the estimated recording medium is a learned recording medium type and a warning icon is not displayed, any paper may be selected by the user from a setting screen (not shown) and associated with a new area and registered.

[0091] As described above, according to this embodiment, the type of recording medium is determined from a combination table of the estimation result of the recording medium and the similarity, and is stored in association with the learned and known type of recording medium. As a result, even if a sheet type is determined to be unlearned, it is possible to identify it as the same recording medium as the known type of recording medium, assuming that it is due to deterioration of the recording medium, lot difference, or environmental change. Also, in this embodiment, as in the example described in the fourth embodiment, even if a sheet type is determined to be unlearned, it is possible to identify it as the same recording medium as the known type of recording medium without reconstructing (re-learning) the learned model.

[0092] <<Other embodiments>> Although the above-described embodiments have been described by taking a recording device as an example of an information processing device, the present invention is not limited to this. The present invention can be applied to any device that estimates the type of recording medium and performs processing according to the estimated type of recording medium. For example, the above-described embodiments may be applied to a scanner that reads an image on a sheet, or a post-processing machine that processes a sheet.

[0093] Furthermore, the estimation of the sheet type and the determination of whether or not it has been learned are not limited to the example in which they are executed by the CPU 301 mounted on the recording apparatus 101, but may be executed by a scanner, a post-processing machine, a PC, or the like.

[0094] The above-described machine learning device may be mounted on the recording device 101. Also, the trained model may be generated in the recording device 101. Teacher data used when generating the trained model may be stored in the memory 305.

[0095] The input data in the teacher data used to determine whether learning has been performed in S505 described above is not limited to the six feature amounts derived in S503 described above, but may also use the color or thickness of the sheet S. Also, it is not limited to a combination of two feature amounts, but one or more feature amounts may be used. Also, the data acquired by the media sensor 206 or the like measured in S502 described above may be used as is as input data without deriving feature amounts.

[0096] The trained model may be stored outside the recording device 101. For example, it may be recorded in a PC or the like connected via the input / output IF 303 and the USB port 304 of the recording device 101. In addition, the trained model is not limited to a DNN, and may be a decision tree or the like.

[0097] In the first and second embodiments described above, for example, in the process of the flowchart in FIG. 5, an example was described in which the characteristic values ​​of the sheet S are derived in S503, and the recording medium is estimated in S504. Then, in S505, whether the type of recording medium (sheet type) estimated in S504 is appropriate is determined using the characteristic values ​​derived in S503. Here, when determining whether the sheet is an unlearned sheet, the estimation process in S504 does not necessarily have to be performed. That is, after each characteristic value is derived in S503, for example, as described in the first embodiment, the characteristic values ​​of each recording medium included in the determination table in FIG. 7 may be compared with the characteristic values ​​derived in S503. Then, when at least one of the characteristic values ​​derived in S503 is outside the range of any recording medium included in the determination table, the CPU 301 may be configured to determine that the sheet S to be processed is an unlearned sheet. That is, when determining whether the sheet is an unlearned sheet, the estimation process in S504 does not necessarily have to be performed. If it is determined that the sheet is not an unlearned sheet, a process corresponding to the estimation process of S504 may be executed. If it is determined that the sheet is an unlearned sheet, the process corresponding to the estimation process of S504 may not be executed, and the result of S506 may be displayed.

[0098] Similarly, in the third embodiment, an example in which the estimation process of S504 is performed and then the process of S505 is performed has been described, but as described in the third embodiment, the estimation process of S505 is performed based on the characteristic value derived in S503. Therefore, also in the third embodiment, the determination process of S505 may be performed without performing the estimation process of S504.

[0099] In the fourth and fifth embodiments, the explanations are based on the example of features described in the first embodiment. However, as described in the second embodiment, the similarity may be derived from the distance between data when two or more features are combined.

[0100] FIG. 18 is a diagram for explaining the combination of the estimation result of the type of sheet and the feature amount. FIG. 18 shows an example in which the similarity is derived from the data distance in a modified example of the fourth embodiment in which two or more feature amounts are combined as described in the second embodiment. FIG. 18(a) shows an example in which a recording medium 4 is obtained as the estimation result of the type of sheet, and the concavity-convexity pair peak 2 in the CIS direction does not fall within the numerical value of the judgment table, and the similarity of the corresponding item is set to 1. FIG. 18(b) shows an example of a combination table in the modified example. FIG. 18(c) shows an example in which the combination of the estimation result and the similarity obtained in FIG. 18(a) is additionally registered as a new recording medium.

[0101] Fig. 19 is a diagram for explaining a combination of the estimation result of the sheet type and the feature amount. Fig. 19 shows an example in which similarity is derived from data distance in a modified example of the fifth embodiment in which two or more feature amounts are combined as described in the second embodiment.

[0102] Also, a form may be formed by selectively combining a part of the fourth embodiment and the fifth embodiment. FIG. 20 is a diagram showing an example of a screen 2000 for accepting a registration method of a recording medium. For example, when the warning icon 851 displayed in S1207 and S1507 is pressed, the CPU 301 may display the screen 2000. The screen 2000 includes a message saying "Please select a registration method of paper." The screen 2000 also includes a "Register new paper" button 2001 for adding the recording medium used in the estimation as a new recording medium, and a "Register to existing paper" button 2002 for linking to an existing recording medium. The screen 2000 may be configured to accept a selection by the user. For example, when the "Register new paper" button 2001 is selected, a process for adding a new type of recording medium to the combination table is performed as described in the fourth embodiment. When the "Register to existing paper" button 2002 is selected, a process for linking to an existing recording medium is performed. In subsequent determinations, by referring to the combination tables of both the fourth and fifth embodiments, if there is a matching combination of estimation results and similarities, the matching type of recording medium can be identified as the type of recording medium used for the estimation.

[0103] Also, another embodiment that partially combines the fourth and fifth embodiments may be adopted. FIG. 21 is a diagram showing an example of a judgment table of characteristic values. FIG. 21 shows a judgment table of characteristic values ​​that are processing switching thresholds, unlike the examples described in FIG. 7 and FIG. 9. FIG. 21(a) is a judgment table corresponding to FIG. 7 described in the first embodiment, and FIG. 21(b) is a judgment table corresponding to FIG. 9 described in the second embodiment. FIG. 21 shows a processing switching threshold set outside the threshold for determining that the recording medium is not yet learned, as described in FIG. 7 and FIG. 9. Then, if there is a characteristic value outside this processing switching threshold, it may be configured to add it as a new recording medium, and if there is no characteristic value outside the threshold, it may be configured to link the area as an existing paper. Alternatively, if the value of 1 of the similarity is a predetermined value or more, it may be configured to add it as a new recording medium, and if it is less than the predetermined value, it may be configured to link the area as an existing paper. Also, these may be appropriately combined.

[0104] Furthermore, in the fourth and fifth embodiments, a flag may be added to the most recently used recording medium for a combination of the estimation result and the similarity. With this configuration, when there are a plurality of recording media with matching combinations of the estimation result and the similarity, the recording medium with the flag set may be preferentially determined.

[0105] Furthermore, the combination of the estimation result and the similarity may be downloaded to a host computer and uploaded to another information processing device. The other information processing device that receives the combination of the estimation result and the similarity may perform the same process as in the above-described embodiment. That is, unlearned sheet information may be shared among a plurality of information processing devices. Furthermore, the table that stores the combination of the estimation result and the similarity may be stored in the information processing device that performs the estimation, or may be stored in another device.

[0106] The present disclosure can also be realized by a process in which a program for implementing one or more functions of the above-described embodiments is supplied to a system or device via a network or a storage medium, and one or more processors in a computer of the system or device read and execute the program. It can also be realized by a circuit (e.g., ASIC) that implements one or more functions.

[0107] The disclosure of the present embodiment includes configurations typified by the following information processing device example, information processing device control method example, and program example.

[0108] <Configuration 1> An acquisition means for acquiring a characteristic value of a recording medium to be fed; a determining means for determining that the recording medium being fed is an unlearned recording medium when the characteristic value acquired by the acquiring means does not correspond to any of a plurality of types of recording media that are determined in advance; 13. An information processing device comprising:

[0109] <Configuration 2> The method further includes estimating a type of the recording medium by using the characteristic value acquired by the acquisition means, The information processing device according to configuration 1, characterized in that the determination means determines that the fed recording medium is an unlearned recording medium when the characteristic value acquired by the acquisition means is not included in a range of characteristics associated with the type of the recording medium estimated by the estimation means.

[0110] <Configuration 3> An acquisition means for acquiring a characteristic value of a recording medium to be fed; an estimation means for estimating a type of the recording medium to be fed from among a plurality of types of recording media determined in advance, using the characteristic value acquired by the acquisition means; a derivation means for deriving a similarity indicating whether or not the characteristic value acquired by the acquisition means is included in a range of characteristics associated with the type of the recording medium estimated by the estimation means; a determination unit that determines the type of the recording medium to be fed based on a combination of the estimation result by the estimation unit and the similarity; 13. An information processing device comprising:

[0111] <Configuration 4> The information processing device according to configuration 3, characterized in that the determination means determines the type of the recording medium to be fed as the type of the recording medium of the matching estimation result based on a match between a combination of the estimation result and the similarity of the recording medium, which is stored in advance in a storage means, and a combination of the estimation result estimated by the estimation means and the similarity derived by the derivation means.

[0112] <Component 5> 5. The information processing device according to configuration 4, further comprising a control means for controlling storage of a combination of the estimation result of the storage medium and the similarity in the storage means as a type of the additional storage medium.

[0113] <Component 6> The information processing device according to configuration 5, wherein when a combination of an estimation result and a similarity of a recording medium, which is stored in advance in the storage means, does not match a combination of an estimation result estimated by the estimation means and a similarity derived by the derivation means, the control means stores the combination of an estimation result estimated by the estimation means and a similarity derived by the derivation means in the storage means as a type of the additional recording medium.

[0114] <Component 7> The information processing device according to configuration 4, further comprising a second control means for controlling the storage of the estimation result in the storage means in addition to the first similarity combined with the estimation result.

[0115] <Component 8> The information processing device according to configuration 7, wherein when a combination of an estimation result and a similarity of a recording medium, which is stored in advance in the storage means, does not match a combination of an estimation result estimated by the estimation means and a similarity derived by the derivation means, the second control means links the estimation result estimated by the estimation means with a second similarity newly derived by the derivation means and stores the linked result in the storage means.

[0116] <Component 9> The information processing device described in configuration 8 is characterized in that even if the combination of the estimation result and similarity of the recording medium previously stored in the memory means does not match the combination of the estimation result estimated by the estimation means and the similarity derived by the derivation means, if the second similarity linked to the estimation result estimated by the estimation means matches, the determination means determines the storage medium with the matching estimation result as the type of the storage medium to be fed.

[0117] <Component 10> a control means for selectively performing a first control for storing a combination of the estimation result and the similarity of the storage medium in the storage means as a type of additional storage medium, and a second control for linking the estimation result to a second similarity in addition to the first similarity combined with the estimation result and storing the second similarity in the storage means; 5. The information processing device according to configuration 4, wherein the control means performs the first control or switches between the second control depending on the degree of similarity.

[0118] <Component 11> a control means for selectively performing a first control for storing a combination of the estimation result and the similarity of the storage medium in the storage means as a type of additional storage medium, and a second control for linking the estimation result to a second similarity in addition to the first similarity combined with the estimation result and storing the second similarity in the storage means; 5. The information processing device according to configuration 4, wherein the control means performs the first control or switches between the second control depending on the characteristic value.

[0119] <Component 12> The storage means; A transmitting means for transmitting the information stored in the storage means to another device; 12. The information processing device according to any one of configurations 4 to 11, further comprising:

[0120] <Component 13> 13. The information processing device according to any one of configurations 4 to 12, further comprising a receiving means for downloading information stored in the storage means of another device to said information processing device.

[0121] <Component 14> The information processing device according to any one of configurations 3 to 13, characterized in that when there are multiple recording media whose combination of the estimation result and similarity of the recording media, which is stored in advance in the storage means, matches the combination of the estimation result estimated by the estimation means and the similarity derived by the derivation means, the determination means determines the recording medium used most recently as the type of recording medium to be fed.

[0122] <Component 15> The information processing device according to any one of configurations 2 to 14, characterized in that the estimation means performs the estimation using a trained model trained based on teacher data including input data corresponding to characteristic values ​​of a recording medium and output data indicating a type of recording medium corresponding to the input data among the plurality of types of recording media determined in advance.

[0123] <Component 16> 16. The information processing apparatus according to any one of configurations 1 to 15, wherein the determining means makes the determination based on information about characteristic values ​​of each of the plurality of predetermined types of recording media.

[0124] <Component 17> the information includes maximum and minimum values ​​for each characteristic value; The information processing device according to configuration 16, characterized in that the determination means determines that the recording medium being fed is an unlearned recording medium if the characteristic value acquired by the acquisition means is not within a range from the maximum value to the minimum value.

[0125] <Component 18> the information includes a range of distances between data when at least two of the characteristic values ​​are combined; The information processing device according to configuration 16, characterized in that the determination means determines that the recording medium being fed is an unlearned recording medium if the distance between data when at least two of the characteristic values ​​acquired by the acquisition means are combined is not included in the range.

[0126] <Component 19> The information processing device of any one of configurations 1 to 15, characterized in that the determination means makes the determination using a second trained model trained based on teacher data including input data corresponding to at least two combinations of characteristic values ​​of the recording medium and output data indicating whether the recording medium is a trained recording medium or an untrained recording medium.

[0127] <Component 20> As the second trained model, a plurality of second trained models are prepared, the number of which corresponds to the combination of the characteristic values; The information processing device described in configuration 19, characterized in that the determination means determines that the fed recording medium is an unlearned recording medium when the result output by at least one of the multiple second trained models indicates that the recording medium is an unlearned recording medium.

[0128] <Component 21> The information processing device according to any one of configurations 1 to 20, characterized in that the acquisition means acquires features relating to surface information of the recording medium and features relating to cross-sectional information of the recording medium, and acquires the characteristic value based on the acquired features.

[0129] <Component 22> The information processing device described in configuration 21, characterized in that the acquisition means acquires features related to the surface information from a sensor that acquires a surface image of the recording medium, and acquires features related to cross-sectional information of the recording medium from a sensor that acquires an electrical signal of ultrasonic waves that have passed through the recording medium.

[0130] <Component 23> 23. The information processing device according to configuration 21 or 22, wherein the surface information includes information regarding at least one of the brightness and the unevenness of the recording medium.

[0131] <Configuration 24> 24. The information processing device according to any one of configurations 21 to 23, wherein the cross-sectional information includes information on at least one of a thickness and a basis weight of the recording medium.

[0132] <Component 25> 25. The information processing device according to any one of configurations 1 to 24, further comprising a notification means for notifying a result determined by the determination means.

[0133] <Component 26> obtaining a characteristic value of a recording medium to be fed; determining that the fed recording medium is an unlearned recording medium when the acquired characteristic value does not correspond to any of a plurality of types of recording media that are determined in advance; 13. A method for controlling an information processing apparatus comprising the steps of:

[0134] <Component 27> A program for causing a computer to execute the control method for an information processing device according to configuration 26. [Explanation of symbols]

[0135] 101 Recording device 206 Media Sensor 207 Ultrasonic Transmitter 301 CPU

Claims

1. An acquisition means for acquiring a characteristic value of a recording medium to be fed; a determining means for determining that the recording medium being fed is an unlearned recording medium when the characteristic value acquired by the acquiring means does not correspond to any of a plurality of types of recording media that are determined in advance; 13. An information processing device comprising:

2. The method further includes estimating a type of the recording medium by using the characteristic value acquired by the acquisition means, 2. The information processing device according to claim 1, wherein the determination means determines that the fed recording medium is an unlearned recording medium when the characteristic value acquired by the acquisition means is not included in a range of characteristics corresponding to the type of recording medium estimated by the estimation means.

3. An acquisition means for acquiring a characteristic value of a recording medium to be fed; an estimation means for estimating a type of the recording medium to be fed from among a plurality of types of recording media determined in advance, using the characteristic value acquired by the acquisition means; a derivation means for deriving a similarity indicating whether or not the characteristic value acquired by the acquisition means is included in a range of characteristics associated with the type of the recording medium estimated by the estimation means; a determination unit that determines the type of the recording medium to be fed based on a combination of the estimation result by the estimation unit and the similarity; 13. An information processing device comprising:

4. 4. The information processing device according to claim 3, wherein the determination means determines the type of the recording medium to be fed as the type of the recording medium of the matching estimation result based on a match between a combination of the estimation result and the similarity of the recording medium stored in advance in a storage means and a combination of the estimation result estimated by the estimation means and the similarity derived by the derivation means.

5. 5. The information processing apparatus according to claim 4, further comprising a control means for controlling storage of a combination of the estimation result of the storage medium and the similarity in the storage means as a type of the additional storage medium.

6. The information processing device according to claim 5, characterized in that when a combination of estimation results and similarities of recording media previously stored in the storage means does not match a combination of the estimation results estimated by the estimation means and the similarities derived by the derivation means, the control means stores the combination of the estimation results estimated by the estimation means and the similarities derived by the derivation means in the storage means as a type of the additional recording medium.

7. 5. The information processing device according to claim 4, further comprising a second control means for controlling the storage of the estimation result in the storage means in addition to the first similarity combined with the estimation result, by linking the estimation result to a second similarity.

8. The information processing device according to claim 7, characterized in that when a combination of the estimation result and similarity of the recording medium previously stored in the storage means does not match a combination of the estimation result estimated by the estimation means and the similarity derived by the derivation means, the second control means links the estimation result estimated by the estimation means to a second similarity newly derived by the derivation means and stores it in the storage means.

9. The information processing device according to claim 8, characterized in that even if the combination of the estimation result and similarity of the recording medium previously stored in the memory means does not match the combination of the estimation result estimated by the estimation means and the similarity derived by the derivation means, if the second similarity linked to the estimation result estimated by the estimation means matches, the determination means determines the storage medium with the matching estimation result as the type of the storage medium to be fed.

10. a control means for selectively performing a first control for storing a combination of the estimation result and the similarity of the storage medium in the storage means as a type of additional storage medium, and a second control for linking the estimation result to a second similarity in addition to the first similarity combined with the estimation result and storing the second similarity in the storage means; The information processing apparatus according to claim 4 , wherein the control means performs either the first control or switches between the second control depending on the degree of similarity.

11. a control means for selectively performing a first control for storing a combination of the estimation result and the similarity of the storage medium in the storage means as a type of additional storage medium, and a second control for linking the estimation result to a second similarity in addition to the first similarity combined with the estimation result and storing the second similarity in the storage means; 5 . The information processing apparatus according to claim 4 , wherein the control means performs either the first control or switches between the second control depending on the characteristic value.

12. The storage means; A transmitting means for transmitting the information stored in the storage means to another device; 5. The information processing apparatus according to claim 4, further comprising:

13. 5. The information processing apparatus according to claim 4, further comprising a receiving means for downloading information stored in said storage means of another apparatus to said information processing apparatus.

14. The information processing device according to claim 3, characterized in that when there are multiple recording media whose combination of the estimation result and similarity of the recording media, which is stored in advance in a storage means, matches the combination of the estimation result estimated by the estimation means and the similarity derived by the derivation means, the determination means determines the recording medium used most recently as the type of recording medium to be fed.

15. The information processing device according to claim 2, characterized in that the estimation means performs the estimation using a trained model trained based on teacher data including input data corresponding to characteristic values ​​of a recording medium and output data indicating the type of recording medium corresponding to the input data among the predetermined plurality of types of recording media.

16. 16. The information processing apparatus according to claim 1, wherein the determining unit makes the determination based on information about characteristic values ​​of each of the plurality of predetermined types of recording media.

17. the information includes maximum and minimum values ​​for each characteristic value; 17. The information processing apparatus according to claim 16, wherein the determining unit determines that the fed recording medium is an unlearned recording medium when the characteristic value acquired by the acquiring unit is not within a range from the maximum value to the minimum value.

18. the information includes a range of distances between data when at least two of the characteristic values ​​are combined; The information processing apparatus according to claim 16, characterized in that the determination means determines that the fed recording medium is an unlearned recording medium if a distance between data when at least two of the characteristic values ​​acquired by the acquisition means are combined is not included in the range.

19. The information processing device according to any one of claims 1 to 15, characterized in that the determination means makes the determination using a second trained model trained based on teacher data including input data corresponding to at least two combinations of characteristic values ​​of the recording medium and output data indicating whether the recording medium is a trained recording medium or an untrained recording medium.

20. As the second trained model, a plurality of second trained models are prepared, the number of which corresponds to the combination of the characteristic values; The information processing device according to claim 19, characterized in that the determination means determines that the fed recording medium is an unlearned recording medium when the result output by at least one of the plurality of second trained models indicates that the recording medium is an unlearned recording medium.

21. 16. The information processing apparatus according to claim 1, wherein the acquisition means acquires a feature amount relating to surface information of the recording medium and a feature amount relating to cross-sectional information of the recording medium, and acquires the characteristic value based on the acquired feature amounts.

22. The information processing device according to claim 21, characterized in that the acquisition means acquires features related to the surface information from a sensor that acquires a surface image of the recording medium, and acquires features related to the cross-sectional information of the recording medium from a sensor that acquires an electrical signal of ultrasonic waves that have passed through the recording medium.

23. 22. The information processing apparatus according to claim 21, wherein the surface information includes information regarding at least one of the brightness and the unevenness of the recording medium.

24. 22. The information processing apparatus according to claim 21, wherein the cross-sectional information includes information regarding at least one of a thickness and a basis weight of the recording medium.

25. 16. The information processing apparatus according to claim 1, further comprising a notification unit that notifies a result determined by said determination unit.

26. obtaining a characteristic value of a recording medium to be fed; determining that the fed recording medium is an unlearned recording medium when the acquired characteristic value does not correspond to any of a plurality of types of recording media that are determined in advance; 13. A method for controlling an information processing apparatus comprising the steps of:

27. A program for causing a computer to execute the method for controlling an information processing device according to claim 26.

Images