Defective product image generation device, defective product image generation method, and program

Abstract

Provided is a defective product image generation device comprising: a data set acquisition unit that acquires one or a plurality of data sets relating to a predetermined expansion target object; a data set designation unit that designates a data set to be used for generating a defective product image from the one or plurality of data sets; a generation property information acquisition unit that acquires generation property information for generating the defective product image; a defective product image generation unit that generates one or a plurality of defective product images from the designated data set on the basis of the generation property information; and a transmission unit that transmits the one or plurality of defective product images, and annotation information corresponding to the one or plurality of defective product images.

Description

Defective product image generating device, defective product image generating method and program

[0001] The present invention relates to a defective product image generating device, a defective product image generating method, and a program.

[0002] It is known to generate image data by applying extended data obtained by extending a characteristic portion of an image to another image (see, for example, Patent Document 1). General disclosure

[0003] In a first aspect of the present invention, a defective product image generation device is provided, comprising: a dataset acquisition unit that acquires one or more datasets related to a predetermined augmented object; a dataset designation unit that designates a dataset to be used to generate a defective product image from the one or more datasets; a generation property information acquisition unit that acquires generation property information for generating the defective product image; a defective product image generation unit that generates one or more defective product images from the designated dataset based on the generation property information; and a transmission unit that transmits the one or more defective product images and annotation information corresponding to the one or more defective product images.

[0004] In the above-mentioned defective product image generation device, the generation property information may include at least one of the type of product that is the augmented object, the image generation model used by the defective product image generation unit, designation information that designates the defective parts to be generated, position information of the defective parts to be generated, morphological information of the defective parts to be generated, the failure mode of the defective parts to be generated, or the number of defective parts to be generated.

[0005] In any of the above-described defective product image generation devices, the transmission unit may transmit the one or more defective product images and one or more generation property values ​​used to generate each of the one or more defective product images.

[0006] Any of the above-mentioned defective product image generation devices may include a storage unit that stores the one or more data sets, the one or more defective product images, and the annotation information corresponding to the one or more defective product images.

[0007] The defective product image generating device may further include a generating user ID acquiring unit configured to acquire a generating user ID of a user who instructs the generation of the defective product image, and the storage unit may store the generating user ID in association with the defective product image generated by the user corresponding to the generating user ID.

[0008] The defective product image generation device may further include a providing user ID acquisition unit that acquires a providing user ID of a user who has provided the one or more data sets. The storage unit may store the providing user ID in association with the one or more data sets provided by the user corresponding to the providing user ID.

[0009] Any of the above-described defective product image generation devices may include a user determination unit that determines a degree of match between a providing user ID of a user who has provided the one or more data sets and a generating user ID of a user who instructs generation of the defective product image. Any of the above-described defective product image generation devices may include a determination unit that determines, based on the degree of match, whether to provide the one or more data sets provided by the user with the providing user ID to the user with the generating user ID.

[0010] Any of the above-mentioned defective product image generation devices may include a user determination unit that determines a degree of match between a providing user ID of a user who provided the one or more data sets and a generating user ID of a user who instructs generation of the defective product image. Any of the above-mentioned defective product image generation devices may also include a warning unit that warns the user of the generating user ID in accordance with the degree of match.

[0011] Any of the above-described defective product image generation devices may include a proposing unit that proposes an image generation model to be used by the defective product image generation unit to generate the defective product image, in accordance with the generation property information.

[0012] Any of the above-described defective product image generating devices may include a classifying unit that classifies the one or more data sets into a plurality of groups divided according to a predetermined viewpoint.

[0013] In any of the above-described defective product image generation devices, the classification unit may classify the one or more data sets by type of product, which is the object to be augmented.

[0014] Any of the above-described defective product image generation devices may include an exclusion property information acquisition unit that acquires exclusion property information to be excluded from a dataset used to generate the defective product image, among the one or more datasets. The defective product image generation unit may generate the defective product image without using a dataset specified by the exclusion property information, among the one or more datasets.

[0015] Any of the above-described defective product image generation devices may include an anonymity flag receiving unit that receives an anonymity flag that specifies whether or not a user who provides the one or more data sets is to be anonymous.

[0016] Any of the above-described defective product image generating devices may include a quality checking unit that checks the quality of the defective product image before the transmitting unit transmits the defective product image.

[0017] In any of the above-described defective product image generating devices, the quality confirmation unit may provide a preview image of the defective product image before the transmission unit transmits the defective product image.

[0018] The defective product image generating device may further include a number specifying unit that specifies the number of defective product images to be acquired by a user, and the transmitting unit may select and transmit the number of defective product images from among defective product images that are greater than the number of images to be acquired.

[0019] Any of the above-described defective product image generating devices may include a reward providing unit that provides a reward to the user based on the disclosure permission information that the user has given permission to be disclosed.

[0020] In a second aspect of the present invention, a method for generating defective product images is provided, comprising the steps of acquiring one or more datasets related to a predetermined augmented object, specifying a dataset to be used to generate defective product images from the one or more datasets, acquiring generation property information for generating the defective product images, generating one or more defective product images from the specified datasets based on the generation property information, and transmitting the one or more defective product images and annotation information corresponding to the one or more defective product images.

[0021] A program is provided to cause a computer to execute the steps of acquiring one or more datasets related to a predetermined augmented object, specifying a dataset from the one or more datasets to be used to generate a defective product image, acquiring generation property information for generating the defective product image, generating one or more defective product images from the specified dataset based on the generation property information, and transmitting the one or more defective product images and annotation information corresponding to the one or more defective product images.

[0022] The above summary of the invention does not list all of the features of the present invention, and subcombinations of these features may also be inventions.

[0023] 2 shows an overview of the configuration of the defective product image generation device 100. Illustrates examples of defective product images and annotation information. Illustrates an example of how a user uses the defective product image generation device 100. Illustrates an example of how a user uses the defective product image generation device 100. Illustrates an example of how a user uses the defective product image generation device 100. Illustrates an example of a modified version of the defective product image generation device 100. Illustrates an example of a modified version of the defective product image generation device 100. Illustrates an example of a modified version of the defective product image generation device 100. Illustrates an example of a modified version of the defective product image generation device 100. Illustrates an example of a modified version of the defective product image generation device 100. Illustrates an example of a modified version of the defective product image generation device 100. Illustrates an example of a modified version of the defective product image generation device 100. Illustrates an example of a computer 2200 in which aspects of the present invention may be embodied in whole or in part.

[0024] The present invention will be described below through embodiments of the invention, but the following embodiments do not limit the scope of the invention as claimed. Furthermore, not all of the combinations of features described in the embodiments are necessarily essential to the solution of the invention.

[0025] FIG. 1 shows an overview of the configuration of a defective product image generation device 100. The defective product image generation device 100 of this example includes a dataset acquisition unit 10, a dataset designation unit 15, a generation property information acquisition unit 20, a defective product image generation unit 25, and a transmission unit 35. The defective product image generation device 100 may also include a model storage unit 40 and a memory unit 45. The illustrated blocks are functionally separated functional blocks and may not necessarily correspond to the actual device configuration. That is, blocks shown as a single block in this figure may not necessarily be configured by a single device. Furthermore, blocks shown as separate blocks in this figure may not necessarily be configured by separate devices. The same applies to blocks in other figures.

[0026] The defective product image generation device 100 generates a defective product image in which a defect point Rd is added to the augmented object based on a predetermined data set related to the augmented object. The augmented object may be an industrial product, a food product, or a daily necessities. However, the type of the augmented object is not limited to these. In one example, the augmented object is a metal product, a pharmaceutical product, a glass product, or an automobile part.

[0027] The defective portion Rd may be a defect such as a flaw, a scratch, a dent, a distortion, noise, a crack, or an attachment. The defective product image generation device 100 may generate a defective product image having one defective portion Rd, or may generate a defective product image having multiple defective portions Rd.

[0028] The dataset acquisition unit 10 acquires one or more datasets related to a predetermined augmented object. The datasets may include a good product dataset and a defective product dataset. The datasets may include a good product dataset and a defective product dataset for each augmented object. The dataset acquisition unit 10 may acquire one or more datasets via a network. The dataset acquisition unit 10 may acquire one or more datasets from a providing user via a network. The dataset acquisition unit 10 may supply the acquired datasets to the defective product image generation unit 25 and the storage unit 45.

[0029] The good-item dataset may include one or more good-item images Ig, which may be images of an augmented object without defects.

[0030] The defective product data set may include one or more pieces of defective product information Id. The defective product information Id may include a data set of defective product images, may include text indicating the name or characteristics of the defective portion Rd, and may include information regarding the failure mode. The text of the defective product information Id may include text describing the defective portion Rd, such as a "deep scratch," "distorted hole," or "dull stain." The defective product information Id may also be a mapping image obtained by image conversion of the defective image.

[0031] The dataset designation unit 15 designates a dataset to be used for generating a defective product image from one or more datasets. The dataset designation unit 15 may designate a dataset according to the augmented object as the dataset to be used for generating a defective product image. For example, the dataset designation unit 15 designates a non-defective product dataset and a defective product dataset according to the augmented object. The dataset designation unit 15 may display a screen for selecting a dataset on a display device, allowing the user to designate an arbitrary dataset from one or more datasets.

[0032] The generation property information acquisition unit 20 acquires generation property information Pd for generating a defective product image. The generation property information acquisition unit 20 may display a screen for specifying the generation property information Pd on a display device, allow a user to specify the generation property information Pd, and acquire the specified generation property information Pd. The generation property information acquisition unit 20 may supply the acquired generation property information Pd to the defective product image generation unit 25.

[0033] The generation property information Pd may include at least one of the type of product that is the extension target, the image generation model used by the defective product image generation unit 25, designation information for designating the defect points Rd to be generated, position information for the defect points Rd to be generated, shape information for the defect points Rd to be generated, the defect mode of the defect points Rd to be generated, or the number of defect points Rd to be generated. However, the generation property information Pd may include other information for generating a defective product image.

[0034] In one example, the product types to be augmented include metal products, pharmaceuticals, glass products, and automobile parts. The product types to be augmented may be further subdivided from the above products. For example, the product types to be augmented include bearings, washers, screws, pills, syringe containers, bottles, engine parts, and automobile bodies. However, the product types to be augmented are not limited to these.

[0035] The image generation model may be a model that has been pre-trained to generate defective product images, such as a Generative Adversarial Network (GAN) or a Diffusion model. Details of the image generation model will be described later.

[0036] The designation information of the defect portion Rd may be information relating to the defect of the defect portion Rd. For example, the designation information includes information on characters and / or images that serve as a base for generating a defective product image.

[0037] The position information of the defective portion Rd may be information about a position where the defective portion Rd is provided in the non-defective image Ig. The position information of the defective portion Rd may be specified in a painted area, may be specified by manually specifying points, or may be specified by a grid with any intervals. The point specification may be performed by specifying the position of the defective portion Rd on the image, or may be performed by specifying the position of the defective portion Rd by coordinates. The position information of the defective portion Rd may be specified in two dimensions or in three dimensions.

[0038] The morphological information of the defective portion Rd may include at least one of the area, major axis, angle, degree of distortion, shape, scratch depth, color, brightness, contrast, and texture information of the defective portion Rd. The morphology of the defective portion Rd may be specified by the magnification of the defective portion Rd, the number of area pixels, the number of major axis pixels, or the like when applying the defective portion Rd to the non-defective image Ig.

[0039] The failure mode of the defective portion Rd may refer to at least one of a defect, a scratch, a dent, a distortion, a noise, a crack, or a deposit at the defective portion Rd, but the types of failure modes of the defective portion Rd are not limited to these.

[0040] The number of defective points Rd is the number of defective points Rd applied to the non-defective image Ig. The number of defective points Rd may be any integer equal to or greater than 1. When the number of defective points Rd is two or more, other generation property information Pd such as designation information of the defective points Rd, position information of the defective points Rd, and / or shape information of the defective points Rd may be acquired for each defective point Rd.

[0041] The generated property information Pd may include a range of property values ​​Vp. Note that the range of the generated property information Pd may refer to the range of property values ​​Vp of the generated property information Pd. Similarly, the range of property values ​​Vp may refer to the range of property values ​​Vp of the generated property information Pd.

[0042] The range of the property value Vp may be a range in which the value of the property value Vp changes continuously. The generated property information Pd may include a maximum value and a minimum value of the property value Vp as the range of the property value Vp. For example, the range of the property value Vp is a continuous range, such as from a minimum value of 10 to a maximum value of 30.

[0043] The range of the property value Vp may be a range in which the value of the property value Vp changes discretely. The generated property information Pd may include a range of the property value Vp specified by setting a predetermined number of divisions or division width for the property value Vp. The generated property information Pd may include a range of the property value Vp specified by setting a maximum and minimum value for the property value Vp and then setting a number of divisions or division width. The generated property information Pd may include a range of the property value Vp specified by a grid. The generated property information Pd may equally divide the predetermined range of the property value Vp by an arbitrary number of divisions and include discrete values ​​corresponding to the number of divisions as the range of the property value Vp. Furthermore, the generated property information Pd may equally divide the predetermined range of the property value Vp by an arbitrary division width and include discrete values ​​corresponding to the division width as the range of the property value Vp. For example, the range of the property value Vp is a discontinuous range such as 10, 20, and 30. The values ​​of the property value Vp may or may not be equally spaced.

[0044] For example, the position information of the defect points Rd may include a range of property values ​​Vp indicating the positions of the defect points Rd to be generated. As another example, the shape information of the defect points Rd may include a range of property values ​​Vp indicating the shapes of the defect points Rd to be generated, and the number of defect points Rd may include a range of property values ​​Vp indicating the number of defect points Rd to be generated.

[0045] The defective product image generation unit 25 generates one or more defective product images from a specified data set based on the generated property information Pd. The defective product image generation unit 25 may generate a defective product image having a defect location Rd according to the generated property information Pd based on the non-defective product image Ig and the defective product information Id included in the data set. If the generated property information Pd includes a range of property values ​​Vp, the defective product image generation unit 25 may generate a defective product image having a defect location Rd according to the range of the property values ​​Vp. In this case, the defective product image generation unit 25 may select an arbitrary property value Vp from the range of the non-defective property values ​​Vp and generate a defective product image according to the selected property value Vp. The defective product image generation unit 25 may supply the generated defective product image to the transmission unit 35 and the storage unit 45.

[0046] The defective product image generation unit 25 may generate a defective product image using an image generation model stored in the model storage unit 40. The defective product image generation unit 25 may select an image generation model corresponding to the generation property information Pd from a plurality of image generation models stored in the model storage unit 40, and generate a defective product image using the selected image generation model. The defective product image generation unit 25 may input a dataset to the image generation model. The image generation model may generate a defective product image in response to the input of the dataset from the defective product image generation unit 25.

[0047] The defective product image generation unit 25 may generate annotation information for one or more defective product images. For each defective product image, the defective product image generation unit 25 may generate annotation information for annotating the defect location Rd in the defective product image. The defective product image generation unit 25 may generate each defective product image and generate annotation information corresponding to each defective product image. That is, the defective product image generation unit 25 may generate annotation information associated with the unwanted product image during the defective product image generation process. The annotation information may include at least one of bounding box information including the position of the defect location Rd, segmented information based on the position and shape of the augmented object corresponding to the defect location Rd, position information of the defect location Rd, or defect mode information. That is, the defective product image generation unit 25 may generate, as annotation information, information from the generation property information Pd that was used to generate the defective product image. The defective product image generation unit 25 may supply the generated annotation information to the transmission unit 35 and the storage unit 45. The entity that generates the annotation information is not limited to the defective product image generating unit 25. The annotation information may be generated by another configuration.

[0048] When annotation information is generated manually by viewing defective product images, it is difficult to add annotation information to all defective product images. In the defective product image generation device 100 of this example, the defective product image generation unit 25 generates annotation information for each defective product image, making it easier to add annotation information compared to manually generating annotation information.

[0049] The transmitting unit 35 transmits one or more defective product images and annotation information corresponding to one or more defective product images. The transmitting unit 35 may transmit the defective product images and the annotation information via a network. The transmitting unit 35 may transmit the defective product images and the annotation information to a receiving user via the network. The transmitting unit 35 may display a screen for selecting a defective product image on a display device, allow the receiving user to select a defective product image to be transmitted, and transmit the selected defective product image and the annotation information corresponding to the defective product image to the receiving user.

[0050] The transmitting unit 35 may transmit one or more defective product images and one or more generated property values ​​used to generate each of the one or more defective product images. The defective product image generating unit 25 of this example generates defective product images based on generated property information Pd. The generated property information Pd may include a range of property values ​​Vp. The generated property values ​​transmitted by the transmitting unit 35 may be property values ​​Vp actually used to generate the defective product image, among the property values ​​Vp included in the range of property values ​​Vp. The transmitting unit 35 may transmit one or more generated property values ​​as annotation information.

[0051] Conventional defective product image generation devices are sometimes installed on-premise in a specific defect detection device and used to generate defective product images for use in that specific defect detection device. The defective product image generation device 100 of this example includes a dataset acquisition unit 10 and a transmission unit 35. This significantly improves convenience in data management and collection, as well as in linking with external systems, compared to defective product image generation devices used on-premise. For example, the defective product image generation device 100 of this example can acquire datasets from multiple providing users, thereby enriching the datasets used to generate defective product images. Furthermore, the defective product image generation device 100 of this example can transmit defective product images generated from datasets of other providing users to receiving users who do not have a dataset for generating defective product images. Furthermore, the defective product image generation device 100 of this example includes a transmission unit 35 that transmits annotation information corresponding to defective product images. Therefore, receiving users can receive annotation information corresponding to defective product images generated from datasets of other providing users. This allows receiving users who receive defective product images that they did not generate to appropriately use the defective product images.

[0052] The model storage unit 40 may store an image generation model for generating a defective product image. The model storage unit 40 may store a plurality of image generation models. For example, the defective product image generation unit 25 inputs a data set to an image generation model stored in the model storage unit 40 and acquires a defective product image generated by the image generation model in response to the input data set. As another example, the model storage unit 40 may supply the defective product image generation unit 25 with an image generation model to be used by the defective product image generation unit 25.

[0053] The storage unit 45 may store one or more data sets, one or more defective product images, and annotation information corresponding to the one or more defective product images. The storage unit 45 may store the data sets, the defective product images, and the annotation information in association with each other.

[0054] The defective product image generation unit 25 may generate a defective product image from a dataset stored in the storage unit 45. For example, the dataset designation unit 15 designates a dataset stored in the storage unit 45. The defective product image generation unit 25 may generate a defective product image from the dataset designated by the dataset designation unit 15. As another example, the dataset designation unit 15 may designate a dataset from the dataset acquired by the dataset acquisition unit 10 and the dataset stored in the storage unit 45, and the defective product image generation unit 25 may generate a defective product image from the dataset designated by the dataset designation unit 15.

[0055] The transmission unit 35 may transmit the defective product image stored in the storage unit 45 and the annotation information stored in the storage unit 45. The transmission unit 35 may also transmit the defective product image generated by the defective product image generation unit 25 and the defective product image stored in the storage unit 45, and the annotation information generated by the defective product image generation unit 25 and the annotation information stored in the storage unit 45.

[0056] The term "dataset acquired by the dataset acquisition unit 10" may also refer to a dataset acquired by the dataset acquisition unit 10 and stored in the storage unit 45. In other words, the dataset acquired by the dataset acquisition unit 10 is a concept that includes not only a dataset that is currently acquired by the dataset acquisition unit 10, but also a dataset that was acquired in advance by the dataset acquisition unit 10. The same applies to defective product images and annotation information.

[0057] 2 shows an example of a defective product image and annotation information. The annotation information may include at least one of information on a bounding box including the position of the defect point Rd, information on segmentation of a position of the augmented object corresponding to the defect point Rd, information on the position of the defect point Rd, or information on the defect mode.

[0058] The annotation information may include a mask image and / or information about the mask image as segmentation information of the positions and shapes of the defect locations Rd. For example, the mask image is a binary image in which areas including the defect locations Rd are displayed in white and other areas are displayed in black. In the mask image of this example, each area displayed in white corresponds to each defect location Rd in the defective product image.

[0059] When there are multiple defect locations Rd, the defective image generating unit 25 may generate annotation information for each defect location Rd. In this example, the number of defect locations Rd is three, so the defective image generating unit 25 generates annotation information for each defect location Rd. The annotation information generated for each defect location Rd may be transmitted and / or stored as one piece of annotation information corresponding to one defective image. In other words, the annotation information corresponding to the defective image may include annotation information for each defect location Rd included in the defective image.

[0060] 3A shows an example of how a user uses the defective product image generation device 100. The defective product image generation device 100 of this example is provided in a server 200.

[0061] User A is a user whose user ID is, for example, aaa. Here, the user ID may be any identifier capable of identifying the user, such as the user's registered email address or username. User A may provide a dataset to the defective product image generation device 100. That is, the dataset acquisition unit 10 may acquire one or more datasets related to a predetermined augmented object from User A.

[0062] The user ID of user A who provides a data set to the defective product image generation device 100 is an example of a providing user ID. In other words, the term providing user ID may refer to the user ID of the user who provided the data set to the defective product image generation device 100.

[0063] User B is a user whose user ID is, for example, bbb. User B may instruct the defective product image generation device 100 to generate a defective product image. The dataset designation unit 15 may cause a display device to display a screen for selecting a dataset, allowing user B to designate an arbitrary dataset from one or more datasets. Similarly, the generation property information acquisition unit 20 may cause a display device to display a screen for designating generation property information Pd, allowing user B to designate generation property information Pd, and acquiring the designated generation property information Pd.

[0064] User B may specify a dataset to be used for generating the defective product image. For example, User B may specify a dataset provided by User A as the dataset to be used for generating the defective product image. User B may specify generation property information Pd for generating the defective product image. The defective product image generation unit 25 may generate one or more defective product images from the dataset specified by User B based on the generation property information Pd specified by User B. The defective product image generation unit 25 may generate annotation information for each of the one or more defective product images generated based on the specification by User B.

[0065] The user ID of user B who instructs the defective product image generation device 100 to generate a defective product image is an example of a generating user ID. In other words, the term generating user ID may refer to the user ID of the user who instructs the defective product image generation device 100 to generate a defective product image.

[0066] User C is a user whose user ID is ccc, for example. User C may receive the defective product image and annotation information from the defective product image generation device 100. That is, the transmission unit 35 may transmit the defective product image and annotation information to User C. The transmission unit 35 may display a screen for selecting a defective product image on the display device, allow User C to select the defective product image to be transmitted, and transmit the selected defective product image and the annotation information corresponding to the defective product image to User C.

[0067] User C may select the defective product image to receive. For example, User C may select the defective product image generated by User B from the data set provided by User A. The transmission unit 35 may transmit to User C the defective product image generated by User B from the data set provided by User A and the annotation information corresponding to the defective product image.

[0068] The user ID of user C who receives the defective product image and annotation information from the defective product image generation device 100 is an example of a receiving user ID. In other words, the term receiving user ID may refer to the user ID of the user who receives the defective product image and annotation information from the defective product image generation device 100.

[0069] As described above, in the defective product image generating device 100 of this example, the user who provides the data set, the user who instructs the generation of the defective product image, and the user who receives the defective product image may be different from one another. Note that different users may mean different individuals, different departments within a company, or different companies. In other words, the user unit may be an individual, a department, or a company.

[0070] In the above example, the description has been given assuming that one user performs each stage, but the number of users performing each stage is not limited to this. The combination of the number of users performing each stage is arbitrary. For example, defective product images may be generated from data sets provided by multiple users, or defective product images may be generated by multiple users, or defective product images may be transmitted to multiple users.

[0071] In this way, the defective product image generation device 100 of this example can significantly improve convenience in data management and collection, as well as in linking with external systems, compared to defective product image generation devices used on-premise. Furthermore, since the defective product image generation device 100 of this example includes a transmission unit 35 that transmits annotation information corresponding to the defective product image, even a receiving user who receives a defective product image that was not generated by the device itself can appropriately use the defective product image.

[0072] 3B shows an example of how users use the defective product image generation device 100. In this example, user A provides a data set to generate a defective product image, and user B receives the defective product image. In this way, at least some of the users who perform each step may be the same.

[0073] 3C shows an example of how users use the defective product image generation device 100. In this example, user A provides a data set, and user B generates and receives a defective product image. As in the example of FIG. 3B, at least some of the users who perform each stage are the same. When at least some of the users who perform each stage are the same, as shown in the examples of FIGS. 3B and 3C, any combination of stages performed by the same user is possible.

[0074] 4 shows a modified example of the defective product image generation device 100. The defective product image generation device 100 of this example differs from the embodiment of FIG. 1 in that it includes a providing user ID acquisition unit 50, a generating user ID acquisition unit 52, a receiving user ID acquisition unit 54, a user determination unit 56, a determination unit 58, and a warning unit 60. In this example, differences from the embodiment of FIG. 1 will be particularly described, and the rest may be the same as the embodiment of FIG. 1.

[0075] The providing user ID acquisition unit 50 acquires the providing user ID of a user who has provided one or more data sets. For example, the providing user ID acquisition unit 50 acquires the user ID of user A who has provided the data sets as the providing user ID. The providing user ID acquisition unit 50 may supply the acquired providing user ID to the storage unit 45 and the user determination unit 56.

[0076] The storage unit 45 may store a providing user ID and one or more datasets provided by a user corresponding to the providing user ID in association with each other. For example, the storage unit 45 stores the user ID of user A in association with a dataset provided by user A. When displaying a screen for selecting a dataset on the display device, the dataset designation unit 15 may display on the display device the dataset and the providing user ID and / or the user corresponding to the providing user ID that is stored in association with the dataset.

[0077] The generating user ID acquisition unit 52 may acquire the generating user ID of the user who instructs the generation of the defective product image. For example, the generating user ID acquisition unit 52 may acquire the user ID of user B who instructs the generation of the defective product image as the generating user ID. The generating user ID acquisition unit 52 may supply the acquired generating user ID to the storage unit 45 and the user determination unit 56.

[0078] The storage unit 45 may store the generating user ID in association with the defective product image generated by the user corresponding to the generating user ID. For example, the storage unit 45 stores the user ID of user B in association with the defective product image generated by user B. When the transmission unit 35 causes the display device to display a screen for selecting a defective product image, the transmission unit 35 may cause the display device to display the defective product image and the generating user ID and / or the user corresponding to the generating user ID that are stored in association with the defective product image.

[0079] The recipient user ID acquisition unit 54 may acquire the recipient user ID of the user who receives the defective product image and the annotation information. For example, the recipient user ID acquisition unit 54 may acquire the user ID of user C who receives the defective product image and the annotation information as the recipient user ID. The recipient user ID acquisition unit 54 may supply the acquired recipient user ID to the storage unit 45 and the user determination unit 56.

[0080] The storage unit 45 may store the recipient user ID and the defective product image received by the user corresponding to the recipient user ID in association with each other. For example, the storage unit 45 stores the user ID of user C in association with the defective product image received by user C.

[0081] The user determination unit 56 may determine the degree of match between the providing user ID of a user who provided one or more data sets and the generating user ID of a user who instructs the generation of a defective product image. The determination unit 58 may determine whether or not to provide one or more data sets provided by the user with the providing user ID to the user with the generating user ID based on the degree of match. For example, if the providing user ID and the generating user ID do not match, the determination unit 58 may determine not to provide the data sets provided by the user with the providing user ID to the user with the generating user ID. However, even if the providing user ID and the generating user ID do not match, the determination unit 58 may determine to provide the data sets provided by the user with the providing user ID to the user with the generating user ID.

[0082] For example, when a user with a providing user ID provides a data set, the user with the providing user ID may supply generating user permission information specifying a generating user ID that is permitted to generate a defective product image from the provided data set to the user determination unit 56. The user determination unit 56 may compare the generating user ID of the user instructing the generation of the defective product image with the generating user permission information and determine the degree of match between the generating user permission information and the generating user ID. The determination unit 58 may determine whether or not to provide the data set provided by the user with the providing user ID to the user with the generating user ID based on the degree of match. For example, when the generating user ID is not included in the generating user permission information, the determination unit 58 may determine not to provide the data set provided by the user with the providing user ID to the user with the generating user ID.

[0083] When it is determined that the data set provided by the user with the providing user ID will not be provided to the user with the generating user ID, the determination unit 58 may control the data set acquisition unit 10 so as not to supply the data set provided by the user with the providing user ID to the defective product image generation unit 25 when the user with the generating user ID generates a defective product image. Alternatively, the determination unit 58 may control the data set designation unit 15 so as not to designate the data set provided by the user with the providing user ID when the user with the generating user ID generates a defective product image, and may control the defective product image generation unit 25 so as not to acquire the data set.

[0084] The warning unit 60 may warn the user of the generating user ID according to the degree of match between the providing user ID and the generating user ID. For example, the warning unit 60 may warn the user of the generating user ID when the providing user ID and the generating user ID do not match. The warning unit 60 may warn the user of the generating user ID according to the degree of match between the generating user permission information and the generating user ID.

[0085] The user determination unit 56 may determine the degree of match between the generating user ID of the user who instructs the generation of the defective product image and the receiving user ID of the user who receives the defective product image and the annotation information. The determination unit 58 may determine whether or not to provide the defective product image generated by the user of the generating user ID to the user of the receiving user ID based on the degree of match. For example, if the generating user ID and the receiving user ID do not match, the determination unit 58 may determine not to provide the defective product image generated by the user of the generating user ID to the user of the receiving user ID. However, even if the generating user ID and the receiving user ID do not match, the determination unit 58 may determine to provide the defective product image generated by the user of the generating user ID to the user of the receiving user ID.

[0086] For example, when a user of a generating user ID instructs the generation of a defective product image, the user of the generating user ID may supply receiving user permission information specifying a receiving user ID that is permitted to receive the generated defective product image to the user determination unit 56. The user determination unit 56 may compare the receiving user ID of the user who will receive the defective product image and annotation information with the receiving user permission information and determine the degree of match between the receiving user permission information and the receiving user ID. The determination unit 58 may determine whether to provide the defective product image generated by the user of the generating user ID to the user of the receiving user ID based on the degree of match. For example, when the receiving user ID is not included in the receiving user permission information, the determination unit 58 may determine not to provide the defective product image generated by the user of the generating user ID to the user of the receiving user ID.

[0087] As another example, when a user with a providing user ID provides a data set, receiving user permission information specifying a receiving user ID that is permitted to receive a defective product image generated from the provided data set may be supplied to the user determination unit 56. The user determination unit 56 may compare the receiving user ID of a user who receives the defective product image and annotation information with the receiving user permission information and determine the degree of match between the receiving user permission information and the receiving user ID. The determination unit 58 may determine, based on the degree of match, whether to provide the user with the receiving user ID with a defective product image generated from the data set provided by the user with the providing user ID.

[0088] When it is determined that the defective product image generated by the user with the generating user ID will not be provided to the user with the receiving user ID, the determination unit 58 may control the transmission unit 35 so as not to transmit the defective product image generated by the user with the generating user ID to the user with the receiving user ID. Alternatively, the determination unit 58 may control the defective product image generation unit 25 so as not to supply the defective product image generated by the user with the generating user ID to the transmission unit 35.

[0089] The warning unit 60 may warn the user of the receiving user ID according to the degree of match between the generating user ID and the receiving user ID. For example, the warning unit 60 may warn the user of the receiving user ID when the generating user ID and the receiving user ID do not match. The warning unit 60 may warn the user of the receiving user ID according to the degree of match between the receiving user permission information and the receiving user ID.

[0090] As described above, the defective product image generating device 100 of this example includes the user determination unit 56, the determination unit 58, and the warning unit 60. As a result, the defective product image generating device 100 of this example can prevent unwanted use and provision while significantly improving convenience in data management and collection, as well as in linking with external systems.

[0091] 5 shows a modified example of the defective product image generating device 100. The defective product image generating device 100 of this example differs from the embodiment of FIG. 1 in that it includes a classification unit 65 and a proposing unit 70. In this example, differences from the embodiment of FIG. 1 will be particularly described, and the rest may be the same as the embodiment of FIG. 1.

[0092] The classification unit 65 may classify one or more datasets into a plurality of groups based on predetermined perspectives. The classification unit 65 may classify datasets into a plurality of groups based on perspectives including at least one of the type of product that is the augmented object, the manufacturing method of the product that is the augmented object, the type of defect, or the type of industry that handles the product that is the augmented object. However, the perspectives by which the classification unit 65 classifies datasets are not limited to these. As an example, the classification unit classifies one or more datasets by the type of product that is the augmented object.

[0093] For example, the data set designation unit 15 may designate a data set included in a group corresponding to the generation property information Pd from among the multiple groups classified by the classification unit 65. The defective product image generation unit 25 may generate a defective product image from a data set included in a group corresponding to the generation property information Pd. The storage unit 45 may store the group used to generate the defective product image in association with the generated defective product image.

[0094] The defective product image generation device 100 of this example includes the classification unit 65, which allows the data set, defective product images, and annotation information to be classified and organized based on predetermined criteria, thereby further improving the convenience of the defective product image generation device 100 in terms of data management and collection, as well as linking with external systems.

[0095] The model storage unit 40 of this example stores a plurality of image generation models 42. The model storage unit 40 of this example stores N image generation models 42. The model storage unit 40 may store a standard model that can be used generally for any augmented object, and a custom model that is individually trained for the defect location Rd. For example, the first image generation model may be the standard model, and the second to Nth image generation models may be custom models.

[0096] The standard model may be a model that can be used generally for any augmented object and has not been individually trained for the defect location Rd. Using the standard model facilitates rapid mass production of data without individual training. For example, if the defect location Rd is a scratch on a washer, the standard model is a model that has not been trained using the same scratch on the washer as the augmented object.

[0097] The custom model may be a model trained individually for the defect location Rd. The custom model is more likely to generate higher quality data than the standard model. For example, if the defect location Rd is a scratch on a washer, the custom model is a model trained using defective product information Id related to the scratch on the washer.

[0098] The multiple image generation models 42 stored in the model storage unit 40 may include custom models corresponding to each of the multiple groups classified by the classification unit 65. In other words, the multiple image generation models 42 stored in the model storage unit 40 may include custom models generated by learning based on the datasets included in each of the multiple groups classified by the classification unit 65.

[0099] The proposing unit 70 may propose an image generation model 42 to be used in generating a defective product image by the defective product image generating unit 25, according to the generation property information Pd. The proposing unit 70 may display a screen for selecting an image generation model 42 on the display device, propose an image generation model 42 to be used in generating a defective product image, and allow the user to select the image generation model 42. As an example, the proposing unit 70 may propose a custom model generated by learning based on a dataset classified by the classifying unit 65 into the same group as the dataset used in generating a defective product image by the defective product image generating unit 25. The proposing unit 70 may propose both a custom model and a standard model.

[0100] The defective product image generation device 100 of this example includes a proposing unit 70, which allows the selection of an appropriate image generation model 42 and improves the accuracy of generating defective product images. This further improves the convenience of data management and collection, as well as linkage with external systems. That is, even when generating defective product images from a data set provided by another user, an appropriate defective product image can be generated.

[0101] 6 shows a modified example of the defective product image generation device 100. The defective product image generation device 100 of this example differs from the embodiment of FIG. 1 in that it includes an excluded property information acquisition unit 75. In this example, differences from the embodiment of FIG. 1 will be particularly described, and the rest may be the same as the embodiment of FIG. 1.

[0102] The exclusion property information acquisition unit 75 may acquire exclusion property information for excluding one or more datasets from the dataset used to generate the defective product image. The defective product image generation unit 25 may generate the defective product image without using the dataset specified in the exclusion property information from the one or more datasets.

[0103] The exclusion property information may include information about metadata included in the dataset. The exclusion property information may include information about the resolution of an image included in the dataset or the number of pixels of a defect location Rd included in the image. For example, if the exclusion property information includes a lower limit value for the image resolution, the defective image generation unit 25 may generate a defective image without using a dataset that includes an image with a resolution lower than the lower limit value.

[0104] The defective product image generating device 100 of this example includes an exclusion property information acquisition unit 75, which allows it to exclude datasets that are inappropriate for generating defective product images. This further improves the convenience of data management and collection, as well as linkage with external systems. That is, even when generating defective product images from datasets provided by other users, it is possible to generate appropriate defective product images.

[0105] 7 shows a modified example of the defective product image generating device 100. The defective product image generating device 100 of this example differs from the embodiment of FIG. 1 in that it includes an anonymous flag receiving unit 80. In this example, differences from the embodiment of FIG. 1 will be particularly described, and the rest may be the same as the embodiment of FIG. 1.

[0106] The anonymity flag receiving unit 80 may receive an anonymity flag that specifies whether a user who provides one or more data sets is to be anonymous. The anonymity flag may specify whether the providing user is to be anonymous to all other users, or may specify that the providing user is to be anonymous to some users.

[0107] For example, when the dataset designation unit 15 displays a screen for selecting a dataset on a display device and allows a user to designate a dataset, the dataset designation unit 15 may display only the dataset on the display device, without displaying the user who provided the dataset on the display device. The dataset designation unit 15 may not display the user who provided the dataset on the display device for some users specified by the anonymous flag, and may display the user who provided the dataset along with the dataset on the display device for other users.

[0108] The defective product image generating device 100 of this example includes an anonymity flag receiving unit 80, which allows the user who provided the data set to remain anonymous from other users. This further improves the convenience of data management and collection, as well as linkage with external systems. In other words, because the user who provided the data set can be kept anonymous, it becomes easier for users who want to provide a data set but do not want their information to be made public to provide the data set.

[0109] Fig. 8 shows a modified example of the defective product image generating device 100. The defective product image generating device 100 of this example differs from the embodiment of Fig. 1 in that it includes a quality confirmation unit 85. In this example, differences from the embodiment of Fig. 1 will be particularly described, and the rest may be the same as the embodiment of Fig. 1.

[0110] The quality confirmation unit 85 may provide information for confirming the quality of the defective product image before the defective product image generation unit 25 generates the defective product image or before the transmission unit 35 transmits the defective product image. The quality confirmation unit 85 may confirm the quality of the defective product image to be generated by checking the defect location Rd before the defective product image is generated, or may confirm the quality of the defective product image by checking the generated defective product image.

[0111] The quality confirmation unit 85 may provide a defect portion preview for checking the defect portion Rd for each piece of generation property information Pd before the defective product image generation unit 25 generates a defective product image. The quality confirmation unit 85 may provide a defect portion preview for checking the shape of the defect portion Rd for each piece of generation property information Pd. For example, if the generation property information Pd includes information such as the defect mode of the defect portion Rd and the size of the defect portion Rd, a user who intends to generate a defective product image can check, via the quality confirmation unit 85, an outline of what kind of defect portion Rd will be generated based on the generation property information Pd.

[0112] The quality confirmation unit 85 may provide a pseudo image generated based on the generation property information Pd to confirm the position and shape of the defect points Rd marked on the augmented object before the defective image generation unit 25 generates a defective image. Here, the pseudo image may be a defective image generated artificially based on the generation property information Pd used to generate the defective image. For example, the pseudo image may be generated artificially without using an image generation model. As an example, the pseudo image may be generated by pasting an image of the defect points Rd on the good product image Ig. As another example, the pseudo image may be generated by pasting a segment of the defect points Rd on the good product image Ig. The method of generating the pseudo image is not particularly limited as long as it generates a pseudo image that allows the position and shape of the defect points Rd marked on the augmented object to be confirmed. A user who intends to generate a defective image can check the artificially generated pseudo image via the quality confirmation unit 85 and then start generating an official defective image using the defective image generation unit 25.

[0113] The quality confirmation unit 85 may provide a preview image of the defective product image before the transmission unit transmits the defective product image. A user who wishes to receive the defective product image can confirm the preview image of the defective product image via the quality confirmation unit 85 and then decide to receive the defective product image.

[0114] As described above, the quality confirmation unit 85 may provide information for confirming the quality of the defective product image before the defective product image generation unit 25 generates the defective product image or before the transmission unit 35 transmits the defective product image. However, the information for confirming the quality of the defective product image provided by the quality confirmation unit 85 is not limited to this.

[0115] The defective product image generating device 100 of this example includes the quality confirmation unit 85, which can further improve convenience in data management and collection, as well as in linking with external systems, etc. That is, even when generating a defective product image from a data set provided by another user, it is possible to generate an appropriate defective product image, and even when receiving a defective product image generated by another user, it is possible to receive an appropriate defective product image.

[0116] 9 shows a modified example of the defective product image generating device 100. The defective product image generating device 100 of this example differs from the embodiment of FIG. 1 in that it includes a number designation unit 90. In this example, differences from the embodiment of FIG. 1 will be particularly described, and the rest may be the same as the embodiment of FIG. 1.

[0117] The number designation unit 90 may designate the number of defective product images to be acquired by the user. The number designation unit 90 may cause a screen for selecting the number of defective product images to be acquired to be displayed on the display device, allowing the user to designate the number of defective product images to be acquired.

[0118] The transmitting unit 35 may select and transmit the acquired number of defective product images from among the defective product images that are greater than the acquired number. The transmitting unit 35 may display a screen for selecting defective product images on the display device, allow the receiving user to select the defective product images to be transmitted, and transmit the selected defective product images to the receiving user. The transmitting unit 35 may display more defective product images than the acquired number on the display device, allow the user to select the acquired number of defective product images. For example, if the acquired number is 10, the transmitting unit 35 may display 100 defective product images on the display device, and allow the user to select 10 defective product images.

[0119] The transmitting unit 35 may select and transmit the number of defective product images with the highest quality from the number of defective product images greater than the number of acquired images. For example, the transmitting unit 35 may select and transmit the number of acquired defective product images in descending order of evaluation based on the evaluation of an evaluation model that evaluates the defective product images generated by the defective product image generating unit 25. In this case, the evaluation model may be a discrimination model using machine learning or a discrimination model that performs rule-based discrimination.

[0120] After the number of images to be acquired is specified, the number designation unit 90 may cause the defective product image generation unit 25 to generate more defective product images than the number of images to be acquired and supply them to the transmission unit 35, or may cause the defective product image generation unit 25 to supply more defective product images than the number of images to be acquired that have been generated in advance to the transmission unit 35.

[0121] The defective product image generating device 100 of this example includes a number designation unit 90 and a transmission unit that selects and transmits a number of defective product images from among defective product images that are greater than the number designated by the number designation unit 90, and therefore can transmit high-quality defective product images preferentially. This further improves convenience in data management and collection, as well as in linking with external systems.

[0122] 10 shows a modified example of the defective product image generation device 100. The defective product image generation device 100 of this example differs from the embodiment of FIG. 1 in that it includes a reward providing unit 95. In this example, differences from the embodiment of FIG. 1 will be particularly described, and the rest may be the same as the embodiment of FIG. 1.

[0123] The reward providing unit 95 may provide a reward to a user based on the disclosure permission information that the user has permitted to be disclosed. The disclosure permission information may include at least one of the number of products that are augmented objects that the user has permitted to be disclosed, the number of defect modes, the number of times the provided dataset has been used by other users, the number of times the generated defective product images have been received by other users, or the degree to which the user has contributed to learning the image generation model 42. However, the types of disclosure permission information are not limited to these. For example, the reward providing unit 95 may provide a higher reward to a user as the number of products that are augmented objects that the user has permitted to be disclosed increases.

[0124] The defective product image generating device 100 of this example includes the reward providing unit 95, which can encourage users to disclose their data. This can further improve the convenience of data management and collection, as well as cooperation with external systems.

[0125] Each functional block illustrated in each of Figures 1 and 4 to 10 may be provided together with functional blocks illustrated in other figures, i.e., the defect image generation device 100 according to the present invention may include all or any combination of the functional blocks illustrated in Figures 1 and 4 to 10.

[0126] 11 illustrates an example of a computer 2200 in which aspects of the present invention may be embodied, in whole or in part. Programs installed on the computer 2200 may cause the computer 2200 to function as or perform operations associated with an apparatus or one or more sections of the apparatus according to embodiments of the present invention, and / or to perform a process or steps of the process according to embodiments of the present invention. Such programs may be executed by the CPU 2212 to cause the computer 2200 to perform specific operations associated with some or all of the blocks of the flowcharts and block diagrams described herein.

[0127] A computer 2200 according to this embodiment includes a CPU 2212, a RAM 2214, a graphics controller 2216, and a display device 2218, which are interconnected by a host controller 2210. The computer 2200 also includes input / output units such as a communication interface 2222, a hard disk drive 2224, a DVD-ROM drive 2226, and an IC card drive, which are connected to the host controller 2210 via an input / output controller 2220. The computer also includes legacy input / output units such as a ROM 2230 and a keyboard 2242, which are connected to the input / output controller 2220 via an input / output chip 2240.

[0128] The CPU 2212 operates according to programs stored in the ROM 2230 and RAM 2214, thereby controlling each unit. The graphics controller 2216 acquires image data generated by the CPU 2212 into a frame buffer or the like provided in the RAM 2214 or into the graphics controller 2216 itself, and causes the image data to be displayed on the display device 2218.

[0129] The communication interface 2222 communicates with other electronic devices via a network. The hard disk drive 2224 stores programs and data used by the CPU 2212 in the computer 2200. The DVD-ROM drive 2226 reads programs or data from the DVD-ROM 2201 and provides the programs or data to the hard disk drive 2224 via the RAM 2214. The IC card drive reads programs and data from an IC card and / or writes programs and data to an IC card.

[0130] ROM 2230 stores therein a boot program or the like that is executed by computer 2200 upon activation, and / or programs that depend on the hardware of computer 2200. I / O chip 2240 may also connect various I / O units to I / O controller 2220 via a parallel port, a serial port, a keyboard port, a mouse port, etc.

[0131] The programs are provided by a computer-readable medium such as a DVD-ROM 2201 or an IC card. The programs are read from the computer-readable medium, installed in the hard disk drive 2224, RAM 2214, or ROM 2230, which are also examples of computer-readable media, and executed by the CPU 2212. Information processing described in these programs is read by the computer 2200, and brings about cooperation between the programs and the various types of hardware resources described above. An apparatus or method may be configured by realizing information manipulation or processing in accordance with the use of the computer 2200.

[0132] For example, when communication is performed between computer 2200 and an external device, CPU 2212 may execute a communication program loaded into RAM 2214 and instruct communication interface 2222 to perform communication processing based on the processing described in the communication program. Under the control of CPU 2212, communication interface 2222 reads transmission data stored in a transmission buffer processing area provided in RAM 2214, hard disk drive 2224, DVD-ROM 2201, or a recording medium such as an IC card, and transmits the read transmission data to the network, or writes received data received from the network to a reception buffer processing area or the like provided on the recording medium.

[0133] Furthermore, the CPU 2212 may cause all or a necessary portion of a file or database stored on an external recording medium such as the hard disk drive 2224, the DVD-ROM drive 2226 (DVD-ROM 2201), an IC card, etc. to be read into the RAM 2214, and may perform various types of processing on the data on the RAM 2214. The CPU 2212 then writes back the processed data to the external recording medium.

[0134] Various types of information, such as various types of programs, data, tables, and databases, may be stored on the recording medium and may undergo information processing. The CPU 2212 may perform various types of processing on data read from the RAM 2214, including various types of operations, information processing, conditional judgment, conditional branching, unconditional branching, information search / replacement, etc., as described throughout this disclosure and specified by the instruction sequences of the programs, and write the results back to the RAM 2214. The CPU 2212 may also search for information in a file, database, etc. on the recording medium. For example, if multiple entries each having an attribute value of a first attribute associated with an attribute value of a second attribute are stored on the recording medium, the CPU 2212 may search for an entry that matches a condition specified by the attribute value of the first attribute from among the multiple entries, read the attribute value of the second attribute stored in the entry, and thereby obtain the attribute value of the second attribute associated with the first attribute that satisfies a predetermined condition.

[0135] The above-described programs or software modules may be stored in a computer-readable medium on or near the computer 2200. A recording medium such as a hard disk or RAM provided in a server system connected to a dedicated communication network or the Internet can also be used as a computer-readable medium, thereby providing the programs to the computer 2200 via the network.

[0136] Although the present invention has been described above using embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments. It will be apparent to those skilled in the art that various modifications and improvements can be made to the above embodiments. It is clear from the claims that such modifications and improvements can also be included within the technical scope of the present invention.

[0137] It should be noted that the order of execution of each process, such as operations, procedures, steps, and stages, in the devices, systems, programs, and methods shown in the claims, specifications, and drawings is not specifically stated as "before," "prior to," etc., and that the processes can be performed in any order unless the output of a previous process is used in a subsequent process. Even if the operational flow in the claims, specifications, and drawings is described using "first," "next," etc. for convenience, this does not mean that the processes must be performed in this order.

[0138] 10... Data set acquisition unit, 15... Data set designation unit, 20... Generated property information acquisition unit, 25... Defective product image generation unit, 35... Transmission unit, 40... Model storage unit, 45... Memory unit, 50... Provided user ID acquisition unit, 52... Generated user ID acquisition unit, 54... Receiving user ID acquisition unit, 56... User judgment unit, 58... Determination unit, 60... Warning unit, 65... Classification unit, 70... Proposal unit, 75... Exclusion property information acquisition unit, 80... Anonymity flag reception unit, 85... Quality confirmation unit, 90... Number designation unit, 100... Defective product image generating device, 200... Server, 2200... Computer, 2201... DVD-ROM, 2210... Host controller, 2212... CPU, 2214... RAM, 2216... Graphics controller, 2218... Display device, 2220... Input / output controller, 2222... Communication interface, 2224... Hard disk drive, 2226... DVD-ROM drive, 2230... ROM, 2240... Input / output chip, 2242... Keyboard

Claims

1. A defective image generation device comprising: a data set acquisition unit that acquires one or more data sets related to a predetermined object to be extended; a data set designation unit that designates a data set to be used for generating a defective image from the one or more data sets; a generation property information acquisition unit that acquires generation property information for generating the defective image; a defective image generation unit that generates one or more defective images from the designated data set based on the generation property information; and a transmission unit that transmits the one or more defective images and annotation information corresponding to the one or more defective images.

2. The defective image generation device according to claim 1, wherein the generation property information includes at least one of a type of the product that is the object to be extended, an image generation model used by the defective image generation unit, designation information for designating a defective part to be generated, position information of the defective part to be generated, morphological information of the defective part to be generated, a defect mode of the defective part to be generated, or the number of defective parts to be generated.

3. The defective image generation device according to claim 1, wherein the transmission unit transmits the one or more defective images and one or more generation property values used for generating each of the one or more defective images.

4. The defective image generation device according to claim 1, further comprising a storage unit that stores the one or more data sets, the one or more defective images, and the annotation information corresponding to the one or more defective images.

5. The defective image generation device according to claim 4, further comprising a generation user ID acquisition unit that acquires a generation user ID of a user who instructs generation of the defective image, wherein the storage unit stores the generation user ID in association with the defective image generated by the user corresponding to the generation user ID.

6. The defective image generation device according to claim 4, further comprising a providing user ID acquisition unit that acquires a providing user ID of a user who provided the one or more data sets, wherein the storage unit stores the providing user ID in association with the one or more data sets provided by the user corresponding to the providing user ID.

7. A user determination unit that determines the degree of coincidence between the providing user ID of the user who provided the one or more data sets and the generating user ID of the user who instructs the generation of the defective product image; and a determination unit that determines whether to provide the one or more data sets provided by the user with the providing user ID to the user with the generating user ID based on the degree of coincidence. The defective product image generation device according to claim 1, comprising:

8. A user determination unit that determines the degree of coincidence between the providing user ID of the user who provided the one or more data sets and the generating user ID of the user who instructs the generation of the defective product image; and a warning unit that warns the user with the generating user ID according to the degree of coincidence. The defective product image generation device according to claim 1, comprising:

9. The defective product image generation device according to claim 1, further comprising a proposal unit that proposes an image generation model to be used for generating the defective product image in the defective product image generation unit according to the generation property information.

10. The defective product image generation device according to claim 1, further comprising a classification unit that classifies the one or more data sets into a plurality of groups divided from a predetermined perspective.

11. The defective product image generation device according to claim 10, wherein the classification unit classifies the one or more data sets for each type of product that is the object to be extended.

12. An exclusion property information acquisition unit that acquires exclusion property information for excluding from the data sets used for generating the defective product image among the one or more data sets; and the defective product image generation unit generates the defective product image without using the data sets specified by the exclusion property information among the one or more data sets. The defective product image generation device according to any one of claims 1 to 11.

13. An anonymous flag reception unit that receives an anonymous flag for designating whether to anonymize the user who provides the one or more data sets. The defective product image generation device according to any one of claims 1 to 11.

14. The defective image generation device according to any one of claims 1 to 11, further comprising a quality confirmation unit for confirming the quality of the defective image before the defective image generation unit generates the defective image or before the transmission unit transmits the defective image.

15. The defective image generation device according to claim 14, wherein the quality confirmation unit provides a defect location preview for confirming a defect location for each generation property information before the defective image generation unit generates the defective image.

16. The defective image generation device according to claim 14, wherein the quality confirmation unit provides a pseudo image generated based on the generation property information in order to confirm the position of a defect location attached to the object to be extended and the form of the defect location before the defective image generation unit generates the defective image.

17. The defective image generation device according to claim 14, wherein the quality confirmation unit provides a preview image of the defective image before the transmission unit transmits the defective image.

18. The defective image generation device according to any one of claims 1 to 11, further comprising a number specification unit for specifying the number of defective images to be acquired by a user, and the transmission unit selects and transmits the defective images of the acquired number from more defective images than the acquired number.

19. The defective image generation device according to any one of claims 1 to 11, further comprising a reward providing unit for providing a reward to the user based on public permission information permitted by the user.

20. A method for generating a defective image, comprising: acquiring one or more data sets related to a predetermined object to be extended; specifying a data set to be used for generating a defective image from the one or more data sets; acquiring generation property information for generating the defective image; generating one or more defective images from the specified data set based on the generation property information; and transmitting the one or more defective images and annotation information corresponding to the one or more defective images.

21. A program for causing a computer to perform the steps of: obtaining one or more data sets related to a predetermined object to be extended; specifying a data set to be used for generating defective product images from the one or more data sets; obtaining generation property information for generating the defective product images; generating one or more defective product images from the specified data set based on the generation property information; and transmitting the one or more defective product images and annotation information corresponding to the one or more defective product images.

Images