Learning program, learning method and learning device

Abstract

To provide a learning program, learning method and learning device capable of improving accuracy of a learned model generated by machine learning using pseudo images.SOLUTION: A learning program causes a computer to execute: an image acquisition step of acquiring a captured non-defective product image and a pseudo-generated defective image including a defective portion; a feature acquisition step of acquiring a first feature of the non-defective product image and a second feature of a region other than the defective portion in the defective image; a determination step of determining whether the first feature and the second feature acquired in the feature acquisition step are similar; and a learning step of executing machine learning of a learning model using the defective image having the second feature when it is determined in the determination step that the first feature and the second feature are similar.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present disclosure relates to a learning program, a learning method, and a learning device. [Background technology]

[0002] Patent Document 1 discloses a learning data generation device that includes a defective part data storage unit that stores defective part data, which is image data of defective part images; a good product data storage unit that stores good product data, which is image data of good product images; a learning data generation unit that generates learning data, which is image data of learning images in which good product images are combined with defective part images, based on the defective part data, good product data, and generation parameters; a learning data storage unit that stores the learning data; and a generation parameter setting unit that sets the generation parameters. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2020-27424 Summary of the Invention [Problem to be solved by the invention]

[0004] One method for inspecting an item involves inputting an image containing the item into a trained model generated by machine learning, and inspecting the item according to the prediction results output from the trained model. In order to improve prediction accuracy, it is preferable to generate a trained model using a large amount of training data. However, in an actual inspection process, defective products are not often generated, making it difficult to obtain many defective images (actual images) of defective products. Therefore, it takes time to obtain many defective images, and therefore it takes time to generate a trained model. Therefore, as described above, pseudo images of defective images are generated, and machine learning of the training model is performed using the pseudo images as training data.

[0005] However, the features of the generated pseudo-images are not necessarily equivalent to the features of the real images actually captured. If there is a large difference between the features of the pseudo-images and the features of the real images, even if machine learning of a learning model is performed using the pseudo-images, the accuracy of the trained model will not improve.

[0006] The present disclosure aims to provide a learning program, a learning method, and a learning device that can improve the accuracy of a trained model generated by machine learning using pseudo images. [Means for solving the problem]

[0007] A learning program according to one embodiment of the present disclosure is a learning program that is executed on at least one processor and performs machine learning on a learning model, and causes a computer to execute an acquisition step of acquiring a pseudo-image generated by simulation, a determination step of determining whether the pseudo-image acquired in the acquisition step satisfies predetermined conditions, and a learning step of performing machine learning on a learning model using the pseudo-image determined in the determination step to satisfy the predetermined conditions. [Effects of the Invention]

[0008] According to the present disclosure, it is possible to improve the accuracy of a trained model generated by machine learning using pseudo images. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram illustrating an example of the hardware configuration of a computer used in a prediction system according to an embodiment. [Figure 2] FIG. 2 is a diagram illustrating an example of a functional configuration of the prediction system. [Figure 3] FIG. 3 illustrates the architecture of a generative adversarial network. [Figure 4] FIG. 4 is a diagram showing an example of the distribution of feature points in a virtual space. [Figure 5] FIG. 5 is a diagram showing an example of the distribution of feature points in a virtual space. [Figure 6] FIG. 6 is a diagram illustrating an example of a neural network. [Figure 7] FIG. 7 is a flowchart showing an example of a learning method and generation of a trained model as a processing flow. [Figure 8] FIG. 8 is a flowchart showing an example of a prediction method as a processing flow. DETAILED DESCRIPTION OF THE INVENTION

[0010] [Description of the embodiments of the present disclosure] First, the contents of the embodiments of the present disclosure will be listed and described. At least some of the embodiments described below may be combined in any combination.

[0011] (1) A learning program according to one aspect of the present disclosure is a learning program that is executed by at least one processor and performs machine learning of a learning model, and causes a computer to execute the following steps: an acquisition step of acquiring a pseudo-image generated by simulation; a determination step of determining whether the pseudo-image acquired in the acquisition step satisfies predetermined conditions; and a learning step of performing machine learning of the learning model using the pseudo-image determined in the determination step to satisfy the predetermined conditions.

[0012] In a learning program according to one aspect of the present disclosure, machine learning of a learning model is performed using pseudo images determined in the determination step to satisfy a predetermined condition. This allows the learning program to exclude pseudo images whose characteristics differ significantly from those of real images, and therefore the pseudo images are not used in the machine learning of the learning model. Therefore, the learning program can perform machine learning of the learning model using pseudo images that have characteristics equivalent to those of real images, thereby improving the accuracy of the trained model generated by machine learning. Therefore, the learning program can improve the accuracy of the trained model generated by machine learning using pseudo images.

[0013] (2) In the learning program of (1) above, the acquisition step may acquire a captured real image, and the determination step may input the pseudo-image and the real image into a generative adversarial network, and determine whether the pseudo-image satisfies a predetermined condition based on the output result of the generative adversarial network. With this configuration, it is possible to accurately determine whether the pseudo-image satisfies the predetermined condition.

[0014] (3) In the learning program of (1) or (2) above, the acquisition step may cause a computer to acquire a captured real image, and in a feature space including the feature points of the pseudo image and the real image, generate a virtual image that may exist between the feature points of the pseudo image and the feature points of the real image. The determination step may input the virtual image and the real image to a generative adversarial network, and determine whether the pseudo image satisfies a predetermined condition based on the output result of the generative adversarial network. This configuration may enable accurate determination of whether the pseudo image satisfies the predetermined condition.

[0015] (4) In any one of the learning programs (1) to (3) above, the acquisition step may cause a computer to acquire a captured real image, and in a feature space including the feature points of the pseudo image and the real image, generate a virtual image that may exist between the feature points of the pseudo image and the feature points of the real image, and the determination step may input each of the pseudo image and the virtual image to a determination model, and determine whether or not the pseudo image satisfies a predetermined condition based on the output result of the determination model. With this configuration, it is possible to accurately determine whether or not the pseudo image satisfies the predetermined condition.

[0016] (5) A learning method according to one aspect of the present disclosure is a learning method executed by at least one processor and performs machine learning of a learning model, and includes an acquisition step of acquiring a pseudo-image generated by simulation, a determination step of determining whether the pseudo-image acquired in the acquisition step satisfies predetermined conditions, and a learning step of performing machine learning of a learning model using the pseudo-image determined in the determination step to satisfy the predetermined conditions.

[0017] In a learning method according to one aspect of the present disclosure, machine learning of a learning model is performed using pseudo images determined in the determination step to satisfy a predetermined condition. This allows the learning method to exclude pseudo images with significantly different characteristics from the real images, and therefore the pseudo images are not used in the machine learning of the learning model. Therefore, the learning method allows machine learning of the learning model using pseudo images with characteristics equivalent to those of the real images, thereby improving the accuracy of the trained model generated by machine learning. Therefore, the learning method improves the accuracy of the trained model generated by machine learning using pseudo images.

[0018] (7) A learning device according to one aspect of the present disclosure is a learning device that includes at least one processor and performs machine learning of a learning model, and includes an acquisition unit that acquires pseudo-images generated by simulation, a determination unit that determines whether the pseudo-images acquired by the acquisition unit satisfy predetermined conditions, and a learning unit that performs machine learning of the learning model using the pseudo-images determined by the determination unit to satisfy the predetermined conditions.

[0019] In a learning device according to one aspect of the present disclosure, machine learning of a learning model is performed using pseudo images determined in the determination step to satisfy a predetermined condition. This allows the learning device to exclude pseudo images whose characteristics differ significantly from those of real images, and therefore the pseudo images are not used in the machine learning of the learning model. Therefore, the learning device can perform machine learning of the learning model using pseudo images that have characteristics equivalent to those of real images, thereby improving the accuracy of the trained model generated by machine learning. Therefore, the learning device can improve the accuracy of the trained model generated by machine learning using pseudo images.

[0020] [Details of the embodiments of the present disclosure] Specific examples of embodiments of the present disclosure will be described below with reference to the drawings. The present disclosure is not limited to these examples, but is defined by the claims, and is intended to include all modifications within the meaning and scope of the claims. In the description of the drawings, the same elements are given the same reference numerals, and duplicate explanations will be omitted.

[0021] [System Overview] The prediction system 1 according to this embodiment is, for example, a computer system that predicts the presence or absence of defects in the inspection of an item. The inspection of the item is, for example, an appearance inspection of the item. In this embodiment, the result predicted by the prediction system 1 is referred to as the "prediction result."

[0022] The prediction system 1 uses machine learning to predict the presence or absence of defects. Machine learning is a technique for autonomously finding laws or rules by iteratively (repeatedly) learning based on given information. The prediction system 1 performs machine learning using a machine learning model. For example, the prediction system 1 may perform machine learning using a gradient boosting decision tree (GBDT), machine learning using multiple regression analysis, or machine learning using a convolutional neural network (CNN) configured to include a convolutional layer and a pooling layer. A convolutional neural network is a type of deep learning that uses a neural network with a multilayer structure.

[0023] The prediction system 1 trains a machine learning model by repeating learning, and acquires this machine learning model as a trained model. This corresponds to the learning phase. In the learning phase, the prediction system 1 functions as a learning device. The trained model is a machine learning model that is predicted to be optimal for predicting the presence or absence of defects. The prediction system 1 processes input data using the trained model and outputs a prediction result of the presence or absence of defects, which corresponds to the operation phase (prediction phase). In the operation phase, the prediction system 1 functions as a prediction device.

[0024] Trained models are portable between computer systems. Therefore, a trained model generated in one computer system can be used in another computer system. Of course, one computer system may execute both the generation and use of trained models. That is, the prediction system 1 may execute both the learning phase and the operation phase, or may not execute either the learning phase or the operation phase. In this embodiment, the prediction system 1 executes both the learning phase and the operation phase.

[0025] In the learning phase, the prediction system 1 uses training data. The training data is image data. The image data may include images of good products and images of defective products. A good product image is an image that does not contain defective parts. A defective image is an image that contains defective parts. The prediction system 1 generates a trained model by performing machine learning using the training data. In the operation phase, the prediction system 1 obtains prediction results by providing input data to the trained model. The input data is image data captured during the inspection process.

[0026] [System Configuration] FIG. 1 is a diagram illustrating an example of a general hardware configuration of a computer 100 constituting the prediction system 1. For example, the computer 100 includes a processor 101, a main memory 102, an auxiliary memory 103, a communication control unit 104, an input device 105, and an output device 106. The processor 101 executes an operating system and application programs. The main memory 102 is composed of, for example, ROM and RAM. The auxiliary memory 103 is composed of, for example, a hard disk or flash memory, and generally stores a larger amount of data than the main memory 102. The communication control unit 104 is composed of, for example, a network card or a wireless communication module. The input device 105 is composed of, for example, a keyboard, a mouse, a touch panel, etc. The output device 106 is composed of, for example, a monitor and a speaker.

[0027] Each functional element of the prediction system 1 is realized by a learning program 110 and a prediction program 120 pre-stored in the auxiliary storage unit 103. Specifically, each functional element is realized by loading the learning program 110 or the prediction program 120 onto the processor 101 or the main storage unit 102 and executing the learning program 110 or the prediction program 120. The processor 101 operates the communication control unit 104, the input device 105, or the output device 106 in accordance with the learning program 110 or the prediction program 120, and reads and writes data from and to the main storage unit 102 or the auxiliary storage unit 103. Data or databases required for processing are stored in the main storage unit 102 or the auxiliary storage unit 103.

[0028] The learning program 110 and the prediction program 120 may be provided by being fixedly recorded on a tangible recording medium such as a CD-ROM, a DVD-ROM, a semiconductor memory, etc. Alternatively, the learning program 110 and the prediction program 120 may be provided via a communication network as a data signal superimposed on a carrier wave.

[0029] The prediction system 1 may be configured with one computer 100 or multiple computers 100. When multiple computers 100 are used, these computers 100 are connected via a communication network such as the Internet or an intranet, thereby logically constructing a single prediction system 1.

[0030] Fig. 2 is a diagram illustrating an example of the functional configuration of the prediction system 1. As shown in Fig. 2, the prediction system 1 includes an image generation unit 10, an image acquisition unit 11, a determination unit 12, a learning unit 13, a storage unit 14, a prediction unit 15, and a database 20.

[0031] The image generation unit 10 is a component that generates a pseudo-image. The image generation unit 10 can access a database that stores image data. The database 20 can be used to train a learning model. The database 20 may be, for example, a component of the prediction system 1, or may be built in a computer system separate from the prediction system 1. The prediction system 1 and the database 20 may be connected via a communication network, or both the prediction system 1 and the database 20 may be built in a single computer.

[0032] The method for preparing the image data to be stored in the database 20 is not limited. For example, the image data may be stored in the database 20 by an operator's input work, or may be automatically collected and stored in the database 20 by the prediction system 1 or another computer system. The database 20 stores images of good products and images of defective products as image data. The images of good products may include real images of good products actually taken in an inspection process or the like, and pseudo-good images generated by the image generation unit 10. The images of defective products may include real images of defective products actually taken in an inspection process or the like, and pseudo-failure images generated by the image generation unit 10.

[0033] The image generating unit 10 acquires an actual defect image from the database 20 and generates a pseudo-defect image. The image generating unit 10 generates the pseudo-defect image by, for example, changing the appearance of a defective part included in the actual defect image or adding noise. The image generating unit 10 generates a pseudo-defect image based on the actual defect image by changing at least one of the shape, size, position, and color of the defective part. The properties of the defective part, such as the shape, size, position, and color, can be set arbitrarily or can be selected randomly. The image generating unit 10 generates the pseudo-defect image using, for example, a neural network. The image generating unit 10 stores the generated pseudo-defect image in the database 20. The image generating unit 10 may acquire a non-defective product image from the database 20 and generate a pseudo-defective product image.

[0034] The image generating unit 10 generates a virtual image based on the actual defect image in the second and third processes described below. The virtual image is a pseudo image. A method for generating the virtual image in the image generating unit 10 will be described later.

[0035] The image acquisition unit 11 is a functional element that acquires image data. The image acquisition unit 11 acquires image data as teacher data (learning data) from the database 20. The image acquisition unit 11 acquires images of good products and images of defective products as image data. The image acquisition unit 11 acquires images of real defective products (hereinafter also simply referred to as "real images") and images of pseudo defective products (hereinafter also simply referred to as "pseudo images") as images of defective products. The image acquisition unit 11 outputs the image data to the determination unit 12.

[0036] The determination unit 12 is a functional element that determines whether or not the pseudo image satisfies predetermined conditions. The determination unit 12 executes at least one of a first process, a second process, and a third process on the pseudo image, and determines whether or not the pseudo image satisfies the predetermined conditions. The determination unit 12 may determine whether or not the pseudo image satisfies the predetermined conditions by one of the first process, the second process, and the third process, or may determine whether or not the pseudo image satisfies the predetermined conditions by any combination of the first process, the second process, and the third process.

[0037] (First treatment) In the first process, the determination unit 12 inputs the pseudo image and the real image into a generative adversarial network (GAN) and determines whether the pseudo image satisfies a predetermined condition based on the output result of the generative adversarial network. FIG. 3 is a diagram showing the architecture of a generative adversarial network. As shown in FIG. 3, the determination unit 12 inputs the pseudo image and the real image into a discriminator in the generative adversarial network and determines whether the pseudo image satisfies a predetermined condition based on the output result of the discriminator. If the output result (class) indicates that the pseudo image is real, the determination unit 12 determines that the pseudo image satisfies the predetermined condition. If the output result indicates that the pseudo image is fake, the determination unit 12 determines that the pseudo image does not satisfy the predetermined condition.

[0038] (Second treatment) In the second process, the determination unit 12 uses the virtual image to determine whether the pseudo image satisfies a predetermined condition. The virtual image is generated by the image generation unit 10. The image generation unit 10 generates a virtual image that may exist between the feature points of the pseudo image and the feature points of the real image in a feature space that includes the feature points of the pseudo image and the real image. The feature points of the pseudo image and the real image can be acquired by inputting each of the pseudo image and the real image into a neural network.

[0039] FIG. 4 is a diagram showing an example of the distribution of feature points in a virtual space. In FIG. 4, feature points of the actual non-defective image are indicated by circles (●), feature points of the actual defective image are indicated by crosses (×), feature points of the pseudo-defective image are indicated by triangles (▲), and feature points of the virtual image are indicated by squares (■). The image generation unit 10 generates a virtual image having a feature point P3 on a virtual line connecting, for example, feature point P1 of the pseudo-defective image and feature point P2 of the actual defective image. The image generation unit 10 generates a virtual image having a feature point P6 on a virtual line connecting, for example, feature point P4 of the pseudo-defective image and feature point P5 of the actual defective image.

[0040] The determination unit 12 inputs the virtual image and the real defective image into the generative adversarial network, and determines whether the pseudo-fault image satisfies a predetermined condition based on the output result of the generative adversarial network. The determination unit 12 inputs the virtual image and the real defective image into a discriminator in the generative adversarial network, and determines whether the virtual image satisfies a predetermined condition based on the output result of the discriminator. If the output result indicates that the virtual image is real, the determination unit 12 determines that the pseudo-fault image satisfies the predetermined condition. If the output result indicates that the virtual image is fake, the determination unit 12 determines that the pseudo-fault image does not satisfy the predetermined condition.

[0041] In the example shown in FIG. 4, the determination unit 12 inputs a virtual image having a feature point P3 to the Discriminator. As shown in FIG. 4, the virtual image having the feature point P3 does not exist in the range R of real images learned by the Discriminator. Therefore, the Discriminator outputs a fake output result. The determination unit 12 determines that the pseudo-failure image having the feature point P1 does not satisfy the predetermined condition because the virtual image is fake.

[0042] In the example shown in FIG. 4, the determination unit 12 inputs a virtual image having a feature point P6 to the Discriminator. As shown in FIG. 4, the virtual image having the feature point P6 exists within the range R of real images learned by the Discriminator. Therefore, a genuine output result is output from the Discriminator. Because the virtual image is genuine, the determination unit 12 determines that the pseudo-fault image having the feature point P4 satisfies a predetermined condition.

[0043] (Third treatment) In the third process, the determination unit 12 uses the virtual image to determine whether the pseudo image satisfies a predetermined condition. The virtual image is generated by the image generation unit 10. The image generation unit 10 generates a virtual image that may exist between the feature points of the pseudo image and the feature points of the real image in a feature space including the feature points of the pseudo image and the real image. The feature points of the pseudo image and the real image can be acquired by inputting each of the pseudo image and the real image into a neural network.

[0044] FIG. 5 is a diagram showing an example of the distribution of feature points in a virtual space. In FIG. 5, feature points of the actual non-defective image are indicated by circles (●), feature points of the actual defective image are indicated by crosses (×), feature points of the pseudo-defective image are indicated by triangles (▲), and feature points of the virtual image are indicated by squares (■). The image generation unit 10 generates a virtual image having a feature point P3 on a virtual line connecting, for example, feature point P1 of the pseudo-defective image and feature point P2 of the actual defective image. The image generation unit 10 generates a virtual image having a feature point P6 on a virtual line connecting, for example, feature point P4 of the pseudo-defective image and feature point P5 of the actual defective image.

[0045] The determination unit 12 inputs the real defective image and the virtual image into a discrimination model, and determines whether the pseudo-defective image satisfies a predetermined condition based on the output result of the discrimination model. The discrimination model is a trained model, and is generated using real defective image and real non-defective image as learning data. The discrimination model is a model that distinguishes between non-defective and defective. If the output results (non-defective, defective) of the discrimination model are the same for the real defective image and the virtual image, the determination unit 12 determines that the pseudo-defective image satisfies the predetermined condition. If the output results of the discrimination model are different for the real defective image and the virtual image, the determination unit 12 determines that the pseudo-defective image does not satisfy the predetermined condition.

[0046] 5, the determination unit 12 inputs a pseudo-fault image having a feature point P2 and a virtual image having a feature point P3 into the discrimination model. As shown in FIG. 5, the feature point P2 of the real fault image and the feature point P3 of the virtual image are present on opposite sides of the discrimination boundary (dashed line) of the discrimination model, and therefore the output result of the discrimination model is different. Therefore, the determination unit 12 determines that the pseudo-fault image having the feature point P2 does not satisfy the predetermined condition.

[0047] 5, the determination unit 12 inputs a pseudo-fault image having a feature point P5 and a virtual image having a feature point P6 into the determination model. As shown in FIG. 5, the feature point P6 of the virtual image exists in the same space as the feature point P3 of the real-fault image with respect to the determination boundary (dashed line) of the determination model, and therefore the output result of the determination model is the same. Therefore, the determination unit 12 determines that the pseudo-fault image having the feature point P5 satisfies the predetermined condition.

[0048] The learning unit 13 is a functional element that performs machine learning. The learning unit 13 performs machine learning of a neural network. FIG. 6 is a diagram showing an example of a neural network NW. As shown in FIG. 6, the neural network NW is composed of, for example, a first layer that is an input layer L1, second, third, and fourth layers that are intermediate layers (hidden layers) L2, and a fifth layer that is an output layer L3. The first layer outputs an input value x=(x0, x1, x2, ... xp) having p parameters as elements to the second layer as is. Each of the second, third, and fourth layers converts the total input into an output using an activation function and passes the output to the next layer. The fifth layer also converts the total input into an output using an activation function, and this output is the output value y=(y0, y1, ... , yp) of the neural network NW having q parameters as elements. q )

[0049] The learning unit 13 performs machine learning of a learning model using images including pseudo-fault images determined by the determination unit 12 to satisfy predetermined conditions. The learning unit 13 can use virtual images for machine learning. The learning unit 13 inputs learning data to the input layer L1 and repeatedly (iteratively) performs a learning process on the neural network NW to generate a trained model. The learning unit 13 stores the trained model obtained by repeating the learning process a predetermined number of times in the storage unit 14.

[0050] The prediction unit 15 is a functional element that predicts the presence or absence of a defect using a trained model. The prediction unit 15 inputs input data to the trained model. In response to inputting the input data to the neural network NW of the trained model, the prediction unit 15 acquires a prediction result including an output value output from the neural network NW.

[0051] In this embodiment, the neural network NW receives the pixel values ​​of each pixel of image data as input data and outputs information indicating a prediction result. The input layer L1 of the neural network NW is provided with neurons equal to the number of pixels in the image. The output layer L3 of the neural network NW is provided with neurons for outputting information related to the prediction result of the target object. The presence or absence of a defect can be predicted based on the output values ​​(likelihoods) of the neurons in the output layer.

[0052] [Operation of the prediction system] (Learning phase) The learning method and the method for generating a trained model will be described with reference to Fig. 7. Fig. 7 is a flowchart showing an example of the learning method and the generation of a trained model as a processing flow S1. The processing flow S1 corresponds to the learning phase.

[0053] In step S11, the image acquisition unit 11 acquires image data from the database 20. In step S12, the determination unit 12 determines whether the pseudo image satisfies predetermined conditions. If the pseudo image satisfies the predetermined conditions (step S12: YES), the learning unit 13 performs machine learning using the pseudo image in step S13. If the pseudo image does not satisfy the predetermined conditions (step S12: NO), the process ends.

[0054] In step S13, the learning unit 13 determines whether to terminate learning. If the termination condition for machine learning is satisfied, the learning unit 13 terminates learning, and if the termination condition is not satisfied, the learning unit 13 continues machine learning. The termination condition is set arbitrarily. For example, the termination condition may be set based on an error, or may be set based on the number of training data to be processed, i.e., the number of times learning is performed.

[0055] If learning is to continue (step S12: NO), the learning unit 13 acquires the next pseudo image and performs the processes from step S13 onwards for that pseudo image. If learning is to end (step S12: YES), the learning unit 13 acquires a trained model in step S14. Thus, in the learning phase, the prediction system 1 generates a trained model by performing machine learning using training data.

[0056] (Operational phase) The prediction method will be described with reference to Fig. 8. Fig. 8 is a flowchart showing an example of the prediction method as a processing flow S2. The processing flow S2 corresponds to the operation phase and is an example of the prediction method according to the present disclosure.

[0057] In step S21, the prediction unit 15 acquires input data. In step S22, the prediction unit 15 inputs the input data to a trained model and outputs a prediction result obtained by the trained model. The method for outputting the prediction result by the prediction unit 15 is not particularly limited. For example, the prediction unit 15 may output the prediction result to the output device 106, store it in a predetermined database, or transmit it to another computer system.

[0058] [effect] As described above, in the prediction system 1 according to this embodiment, machine learning of a learning model is performed using pseudo images determined by the determination unit 12 to satisfy predetermined conditions. This allows the prediction system 1 to exclude pseudo images whose characteristics are significantly different from those of the real image, and therefore the pseudo images are not used in the machine learning of the learning model. Therefore, the prediction system 1 can perform machine learning of a learning model using pseudo images that have characteristics equivalent to those of the real image, thereby improving the accuracy of the trained model generated by machine learning. Therefore, the prediction system 1 can improve the accuracy of the trained model generated by machine learning using pseudo images.

[0059] In the prediction system 1 according to this embodiment, in the first process, the determination unit 12 inputs a pseudo image and a real image into a generative adversarial network, and determines whether the pseudo image satisfies a predetermined condition based on the output result of the generative adversarial network. With this configuration, it is possible to accurately determine whether the pseudo image satisfies the predetermined condition.

[0060] In the prediction system 1 according to this embodiment, in the second process, the determination unit 12 inputs the virtual image and the real image into the generative adversarial network and determines whether the pseudo image satisfies a predetermined condition based on the output result of the generative adversarial network. With this configuration, it is possible to accurately determine whether the pseudo image satisfies the predetermined condition.

[0061] In the prediction system 1 according to this embodiment, in the third process, the determination unit 12 inputs each of the pseudo image and the virtual image into a determination model, and determines whether the pseudo image satisfies a predetermined condition based on the output result of the determination model. With this configuration, it is possible to accurately determine whether the pseudo image satisfies the predetermined condition.

[0062] Although the embodiments of the present disclosure have been described above, the present disclosure is not necessarily limited to the above-described embodiments, and various modifications are possible without departing from the spirit of the present disclosure.

[0063] In the above embodiment, the image generating unit 10 generates the pseudo-fault image using a neural network as an example. However, the method of generating the pseudo-fault image by the image generating unit 10 is not limited to this.

[0064] In the above embodiment, the neural network NW is described as being composed of a first layer, which is an input layer L1, a second layer, a third layer, and a fourth layer, which are intermediate layers (hidden layers) L2, and a fifth layer, which is an output layer L3. However, the neural network may include one or more intermediate layers. [Explanation of symbols]

[0065] 1. Prediction system 10...Image generation unit 11...Image acquisition unit 12...Judgment section 13...Study Department 14...Storage section 15...Prediction Department 20...Database 100...Computer 101...Processor 102...Main memory section 103…Auxiliary storage unit 104...Communication control unit 105...Input device 106...Output device 110...Study Program 120...Prediction Program L1...input layer L2: Intermediate layer (hidden layer) L3: Output layer NW...neural network P1: Minutiae P2: Minutiae P3: Minutiae P4: Minutiae P5: Minutiae P6…Feature points S1...Processing flow S2...Processing flow S11...Step S12...Step S13...Step S14...Step S21...Step S22...Step

Claims

1. A learning program that is executed on at least one processor and performs machine learning of a learning model, an acquisition step of acquiring a pseudo-generated pseudo image; a determination step of determining whether the pseudo image acquired in the acquisition step satisfies a predetermined condition; a learning step of performing machine learning of the learning model using the pseudo image determined to satisfy the predetermined condition in the determination step; A learning program that causes a computer to execute the following.

2. In the acquiring step, a captured actual image is acquired, The learning program according to claim 1 , wherein the determination step inputs the pseudo image and the real image into a generative adversarial network, and determines whether the pseudo image satisfies the predetermined condition based on the output result of the generative adversarial network.

3. In the acquiring step, a captured actual image is acquired, a generating step of generating a virtual image that may exist between the feature points of the pseudo image and the feature points of the real image in a feature space including the feature points of the pseudo image and the real image, 2. The learning program according to claim 1, wherein the determining step inputs the virtual image and the real image into a generative adversarial network, and determines whether the pseudo-image satisfies the predetermined condition based on the output result of the generative adversarial network.

4. In the acquiring step, a captured actual image is acquired, a generating step of generating a virtual image that may exist between the feature points of the pseudo image and the feature points of the real image in a feature space including the feature points of the pseudo image and the real image, 2. The learning program according to claim 1, wherein in the judgment step, the pseudo image and the virtual image are each input into a judgment model, and it is judged whether or not the pseudo image satisfies the predetermined condition based on the output result of the judgment model.

5. A learning method for performing machine learning on a learning model, the method being executed on at least one processor, comprising: an acquisition step of acquiring a pseudo-generated pseudo image; a determination step of determining whether the pseudo image acquired in the acquisition step satisfies a predetermined condition; A learning step of performing machine learning on the learning model using the pseudo-images determined to satisfy the predetermined conditions in the determination step.

6. A learning device that includes at least one processor and performs machine learning on a learning model, an acquisition unit that acquires a pseudo image that is generated in a pseudo manner; a determination unit that determines whether the pseudo image acquired by the acquisition unit satisfies a predetermined condition; A learning device comprising: a learning unit that performs machine learning on the learning model using the pseudo image that is determined by the determination unit to satisfy the specified condition.

Images