Evaluation system, evaluation method, and evaluation program

Abstract

This evaluation system is provided with at least one processor. The at least one processor performs: a process for acquiring an object image representing an object having a substrate and a coating region on the substrate; inputting the object image into a learning completion model for estimating a coating area from the input image, and determining the coating area of the object; and calculating an evaluation value relating to the specified coating region.

Description

Technical Field

[0001] One aspect of the present invention relates to an evaluation system, an evaluation method, and an evaluation program. Background Art

[0002] Patent Document 1 describes a conductive particle shape evaluation device that evaluates the surface shape of a conductive particle having a plurality of conductive protrusions on its surface.

[0003] Previous technical literature

[0004] Patent Literature

[0005] Patent Document 1: Japanese Patent Application Laid-Open No. 2016-061722 Summary of the Invention

[0006] Technical issues to be solved by the invention

[0007] A mechanism for easily performing evaluation related to coverage of an object is desired.

[0008] Means for solving technical problems

[0009] An evaluation system according to one aspect of the present invention includes at least one processor. The at least one processor performs the following processing: acquiring an object image representing an object having a substrate and a coating region on the substrate; inputting the object image into a learned model for estimating the coating region based on the input image to determine the coating region of the object; and calculating an evaluation value associated with the determined coating region.

[0010] An evaluation method according to one aspect of the present invention is performed by an evaluation system including at least one processor. The evaluation method includes the following steps: acquiring an object image representing an object having a substrate and a coating region on the substrate; inputting the object image into a learned model for estimating the coating region based on the input image to determine the coating region of the object; and calculating an evaluation value associated with the determined coating region.

[0011] An evaluation program according to one aspect of the present invention causes a computer to execute the following steps: obtaining an object image representing an object having a substrate and a coating area on the substrate; inputting the object image into a learned model for estimating the coating area based on the input image to determine the coating area of the object; and calculating an evaluation value related to the determined coating area.

[0012] In such an aspect, the coating region of the object is estimated from the object image using the learned model. With this structure, the coating region can be easily identified, and the calculation of the evaluation value related to the coating region also becomes easy.

[0013] Therefore, evaluation regarding coverage of the object can be easily performed.

[0014] Effects of the Invention

[0015] According to one aspect of the present invention, evaluation regarding coverage of an object can be easily performed. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a diagram showing the functional structure of the evaluation system.

[0017] Figure 2 This is a flowchart showing an example of generating an object image.

[0018] Figure 3 Is to express Figure 2 Figure related to image processing examples.

[0019] Figure 4 Is to express Figure 2 Figure related to image processing examples.

[0020] Figure 5 This is a flowchart showing an example of generating a teacher image.

[0021] Figure 6 Is to express Figure 5 Figure related to image processing examples.

[0022] Figure 7 This is a flowchart showing an example of generating a learned model.

[0023] Figure 8 This is a flowchart showing an example of evaluation related to coverage of an object.

[0024] Figure 9 Is to express Figure 8 Figure 1. Related image processing examples. DETAILED DESCRIPTION

[0025] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the description of the drawings, the same or equivalent elements are denoted by the same reference numerals, and repeated descriptions are omitted.

[0026] [System Overview]

[0027] The evaluation system involved in the present invention is a computer system that performs an evaluation related to the coverage of an object projected on an image. The object has a substrate and a coating area. Coverage refers to a state in which at least a portion of the substrate is hidden by the coating area in the appearance of the object. The evaluation system performs an evaluation related to the coating area, for example, an evaluation related to the coverage of the substrate based on the coating area. The evaluation related to the coating area refers to a process of quantitatively judging the coating area covering the substrate. In one example, the evaluation system calculates a quantitative indicator related to the coating area, i.e., an evaluation value, and outputs the evaluation value. For example, the evaluation value is a value related to the coverage of the substrate based on the coating area.

[0028] The object is a solid object to be evaluated by the evaluation system. The substrate is a component that occupies the main area of the object. The coating area is a component located on the substrate. The object has any shape, size, and composition. For example, the object can be spherical, flat, or cylindrical, or it can have a more complex shape. The object can be of a size that can be visually identified, or it can be so small that it cannot be confirmed without a microscope. The shape and size of the substrate can affect the shape and size of the object. The coating area covers at least a portion of the surface of the substrate. The coating area can be formed by attaching powdered or granular particles to the substrate, or by applying a liquid coating agent or a coating agent containing a metal to the substrate. Multiple coating areas separated from each other can be provided on a single substrate. The coating area can be formed by multiple coating elements arranged on the substrate. As an example of a coating element, individual particles can be cited. Particles arranged on the substrate as coating elements can be understood as protrusions. The coating area can be understood as a convex portion that is raised above the surface of the substrate, or it can be understood as a collection of dense protrusions. The base material and the coating region may differ in at least a portion of their components, or may have all the same components. Both the base material and the coating region may be organic compounds, inorganic compounds, or both.

[0029] The object can be a particulate material. In one example, the particulate material comprises a core particle and a plurality of microparticles disposed on the surface of the core particle. The diameter of the microparticles is smaller than the diameter of the core particle. The core particle is an example of a substrate, each microparticle is an example of a coating element, and the collection of the microparticles is an example of a coating region.

[0030] The evaluation system uses a learned model generated by machine learning to determine the coating area of the object. The learned model is a computational model that infers the coating area of the object based on an image representing the object. Machine learning refers to a method of autonomously searching for patterns or rules by repeatedly learning based on the provided information. The evaluation system inputs the image of the object into the learned model, determines the coating area, and performs an evaluation related to the determined coating area. The evaluation system can generate a learned model through machine learning. The generation of the learned model is equivalent to the learning stage, and the determination of the coating area based on the learned model is equivalent to the application stage. The learning model used in the evaluation system can be a model that can perform instance segmentation or panoptic segmentation, for example, Mask R-CNN, DeepMask, FCIS (Fully Convolutional Instance-aware Semantic Segmentation), Panoptic Feature Pyramid Network or UPSNet.

[0031] By importing the learned model, the coating area can be automatically and accurately determined. Depending on the properties of the object and the coating area, it is sometimes necessary for a person to visually determine the coating area. There is the following prior art: the image of the object is converted into a binary image that distinguishes the coating area from other areas, and the coating area is determined based on the binary image. However, in this prior art, it is necessary for a person to visually set a binarization threshold for each image, and the processing takes time. In addition, the determination of the coating area may rely on human perception. Moreover, it is not easy to accurately determine the coating area. In view of this prior art, the evaluation system can automatically and accurately determine the coating area, so the user of the evaluation system can simply obtain an evaluation related to the coverage of the object.

[0032] [System Structure]

[0033] Figure 1This figure shows the functional structure of an evaluation system 10 according to an example. The evaluation system 10 includes a processor 101 as a hardware component. The processor 101 is, for example, a CPU (Central Processing Unit), a DSP (Digital Signal Processor), or a GPU (Graphics Processing Unit). The evaluation system 10 also includes, as hardware components: a main storage device composed of RAM and ROM; an auxiliary storage device composed of flash memory, a hard disk, etc.; input devices such as a keyboard and a mouse; output devices such as a monitor and speakers; and a communication module for performing data communications with external devices. The various functional modules of the evaluation system 10 are implemented by the processor 101 executing programs stored in the auxiliary storage device.

[0034] The evaluation program for causing the computer to function as the evaluation system 10 includes program code for implementing the various functional modules of the evaluation system 10. This evaluation program can be provided on a non-transitory recording medium such as a CD-ROM, DVD-ROM, or semiconductor memory. Alternatively, the evaluation program can be provided via a communication network as a data signal superimposed on a carrier wave. The provided evaluation program can be stored, for example, in an auxiliary storage device.

[0035] The evaluation system 10 can consist of a single computer or a collection of multiple computers, i.e., a distributed system. Examples of computers used in the evaluation system 10 include various types of computers, such as personal computers, workstations, tablet computers, and smartphones. When multiple computers are used for the evaluation system 10, these computers are connected via a communication network such as the Internet or an intranet, thereby logically forming a single evaluation system 10. The evaluation system 10 can be implemented as a client-server system, such as a cloud system, or using standalone computers.

[0036] In one example, the evaluation system 10 is connected to at least one external storage device via a communication network. The external storage device is a device or recording medium that stores various data used in the processing of the evaluation system 10. The external storage device may be a component of the evaluation system 10 or may be located in a computer system separate from the evaluation system 10. The communication network may be constructed using the Internet, an intranet, or a combination thereof. The communication network may be constructed using a wired network, a wireless network, or a combination thereof.

[0037] Figure 1, a first original image database 31, a teacher image database 32, and a second original image database 33 are shown as examples of external storage devices. The first original image database 31 and the second original image database 33 are each storage devices that store at least one original image representing at least one object. The teacher image database 32 is a storage device that stores at least one teacher image used for machine learning. At least two of the first original image database 31, the teacher image database 32, and the second original image database 33 can be integrated into a single database.

[0038] In one example, the processor 101 functions as a preprocessing unit 11, a learning unit 12, and an evaluation unit 13. The preprocessing unit 11 is used in both the learning phase and the operational phase. The learning unit 12 corresponds to the learning phase, and the evaluation unit 13 corresponds to the operational phase.

[0039] The preprocessing unit 11 is a functional module that generates an object image representing a single object. During the learning phase, the preprocessing unit 11 generates a sample image 42, an example of an object image, from a first original image 41 read from the first original image database 31. Sample image 42 is used to generate a teacher image 43. During the operational phase, the preprocessing unit 11 generates an object image 45, another example of an object image, from a second original image 44 read from the second original image database 33. Object image 45 is used to evaluate the coverage of the object.

[0040] The learning unit 12 is a functional module that generates the learned model 20. In one example, the learning unit 12 includes a training image generation unit 121 and a model generation unit 122. The training image generation unit 121 is a functional module that generates a training image 43 based on a sample image 42. The model generation unit 122 is a functional module that generates the learned model 20 through machine learning based on the training image 43.

[0041] The evaluation unit 13 is a functional module that performs an evaluation related to the coating of an object. In one example, the evaluation unit 13 includes a coating identification unit 131 and a calculation unit 132. The coating identification unit 131 is a functional module that uses the learned model 20 to identify the coating area of the object based on the object image 45. The calculation unit 132 is a functional module that calculates an evaluation value based on the identified coating area.

[0042] [System Action]

[0043] The following describes an example of processing by the evaluation system 10 and an example of the evaluation method according to the present invention. In the following example, a particulate material is shown as an object, and each fine particle on a core particle is shown as a coating element.

[0044] (Generation of Object Image)

[0045] refer to Figures 2 to 4 , the generation of the object image is explained. Figure 2 An example of such processing is shown in a flowchart as processing flow S1. Figure 3 and Figure 4 1 is a diagram showing an example of image processing related to the processing flow S1. The evaluation system 10 can execute the processing flow S1 in both the learning phase and the operation phase.

[0046] In step S11, the preprocessing unit 11 obtains an original image. In one example, the original image representing the particulate material is an SEM image obtained by photographing a plurality of particulate materials collected on a carbon ribbon using a scanning electron microscope (SEM). In the case where auxiliary information such as character strings and scales are marked in the original image, the preprocessing unit 11 can remove the auxiliary information through image processing such as cropping to obtain an original image that does not include the auxiliary information. The preprocessing unit 11 reads the first original image 41 from the first original image database 31 in the learning phase and reads the second original image 44 from the second original image database 33 in the application phase. In the following description related to the processing flow S1, the first original image 41 and the second original image 44 are collectively referred to as "original images."

[0047] In step S12, the preprocessing unit 11 converts the original image into a binary image. In one example, the preprocessing unit 11 applies an averaging filter based on a predetermined kernel size to the original image to blur the original image. Next, the preprocessing unit 11 performs a binarization process such as Otsu binarization on the blurred original image, and converts the brightness of the object area, which is the area of the object, to 255, and converts the brightness of the background area to 0. As a result, a binary image is generated in which the object is represented by white and the background is represented by black. In this binary image, if there are black spots as noise in the white area, the preprocessing unit 11 removes the black spots. By applying the averaging filter to the original image before the binarization process, it is possible to prevent or suppress the generation of multiple black spots in the area of the object. Figure 3 It shows that the original image 201 is converted into a binary image 202 .

[0048] In step S13, the preprocessing unit 11 determines the center of each object by distance conversion. The distance conversion of the binary image or grayscale image is processed as follows: for each pixel whose brightness is not 0, the distance to the nearest pixel with brightness 0 is calculated, and a distance map representing each distance is generated. The distance map represents each distance in grayscale in such a way that the brightness increases as the distance increases. For each white pixel in the object area, the preprocessing unit 11 calculates the distance to the nearest black pixel in the background area. Based on the distance of each white pixel, the preprocessing unit 11 generates a distance map corresponding to the binary image. Based on the distance map, the preprocessing unit 11 extracts the area with brightness above a specified threshold as the central part of each object. The preprocessing unit 11 determines the pixel with the highest pixel value in each central part as the center of the object. Figure 3 The binary image 202 is converted into a distance map 203. The central portions with relatively high brightness in the distance map 203 correspond to the objects in the original image 201. The pixel with the highest brightness in each central portion of the distance map 203 represents the center of the object.

[0049] In step S14, the pre-processing unit 11 generates a reference image showing the center portion of each object. As described above, the center portion is obtained from the distance map. Figure 3 The reference image 204 is generated based on the distance map 203. The reference image 204 is represented by white in the center and black in the other areas. In one example, the pre-processing unit 11 displays the reference image on a monitor. The user can confirm the generation of the object image through the reference image.

[0050] In step S15, the preprocessing unit 11 selects an object that is reflected as a whole in the original image. In one example, the preprocessing unit 11 determines the size of the object based on the distance map and the distance set at the pixel corresponding to the center of the object. Then, the preprocessing unit 11 assumes the shape of the object based on the center and size of the object. The preprocessing unit 11 selects the object whose entire shape is located within the original image as the object that is reflected as a whole in the original image. For example, the preprocessing unit 11 obtains the distance set to the pixel corresponding to the center of the object as the radius of the object. Then, the preprocessing unit 11 assumes the virtual circle defined by the center and radius as the shape of the object. The preprocessing unit 11 selects the object whose entire shape is located within the original image as the object that is reflected as a whole in the original image. Figure 4 The image 205 shown shows that seven particulate matter appearing entirely in the original image 201 have been selected based on the distance map 203 .

[0051] In step S16, the pre-processing unit 11 extracts the selected object from the original image to generate an object image. The pre-processing unit 11 generates an object image for each of the n selected objects, and as a result, obtains n object images. Figure 4 The image group 206 shown is seven object images corresponding to the seven objects selected from the original image 201. The object images are sample images 42 in the learning phase and object images 45 in the application phase. In the learning phase, the preprocessing unit 11 generates one or more sample images 42 from a first original image 41. The preprocessing unit 11 may store the sample images 42 in a predetermined storage device or output the sample images 42 to the learning unit 12 for generating a teacher image 43. In the application phase, the preprocessing unit 11 generates one or more object images 45 from a second original image 44. The preprocessing unit 11 may store the object images 45 in a predetermined storage device or output the object images 45 to the evaluation unit 13 for evaluation related to the coverage of the objects.

[0052] The evaluation system 10 may execute the processing flow S1 multiple times or repeatedly in each of the learning phase and the operational phase.

[0053] (Generation of teacher images)

[0054] refer to Figure 5 and Figure 6 , the generation of teacher images is explained. Figure 5 An example of such processing is shown in a flowchart as processing flow S2. Figure 6 1 is a diagram showing an example of image processing related to the processing flow S2. The processing flow S2 corresponds to the generation of training data used in the learning phase.

[0055] In step S21, the teacher image generating unit 121 displays the sample image 42 to be processed on the monitor. For example, the teacher image generating unit 121 may display the sample image 42 selected by the user or the sample image 42 input from the pre-processing unit 11.

[0056] In step S22, the training image generation unit 121 receives input of labels for the sample image 42. Labels are information that is treated as ground truth in machine learning. The labels for the sample image 42 represent coating regions, for example, regions representing each coating element. In one example, the training image generation unit 121 receives labels set based on user input.

[0057] In step S23, the teacher image generation unit 121 generates a teacher image 43 based on the sample image 42 and the label. For example, the teacher image generation unit 121 may embed the label in the sample image 42 to generate the teacher image 43. The teacher image generation unit 121 stores the generated teacher image 43 in the teacher image database 32. Figure 6 This figure shows the generation of a teacher image 220 based on sample image 210. Teacher image 220 includes labels representing the areas of each coating element that forms the coating area of the object. Each label can be displayed using various representation methods, such as color, pattern, or mark, to distinguish each coating element. In the example shown as teacher image 220, the fact that two or more coating elements have the same pattern does not indicate a relationship between the coating elements, but rather indicates that the coating elements are individually identified.

[0058] The evaluation system 10 can execute the processing flow S2 multiple times or repeatedly. The evaluation system 10 executes the processing flow S2 for each sample image 42 , thereby accumulating the teacher image 43 in the teacher image database 32 .

[0059] (Generation of the learning model)

[0060] refer to Figure 7 , explaining the generation of the learned model. Figure 7 1 is a flowchart showing an example of the processing as a processing flow S3. The processing flow S3 corresponds to the learning phase.

[0061] In step S31 , the model generation unit 122 acquires one teacher image 43 from the teacher image database 32 .

[0062] In step S32, the model generation unit 122 performs learning based on the teacher image 43. For example, the model generation unit 122 performs learning based on Mask R-CNN. In one example, the model generation unit 122 inputs the teacher image 43 into a machine learning model including a neural network, and obtains an estimated result of the coating area output from the machine learning model. The model generation unit 122 updates the parameters in the machine learning model based on the error between the estimated result and the label of the teacher image 43, and using methods such as back propagation (error back propagation method). For example, the model generation unit 122 updates the weights of the neural network.

[0063] In step S33, it is determined whether the model generation unit 122 has ended the machine learning. When the model generation unit 122 determines that the prescribed end condition is not met ("No" in step S33), the process returns to step S31. In the repeated process, the model generation unit 122 obtains the next teacher image 43 in step S31, and performs learning based on the teacher image 43 in step S32. On the other hand, when the model generation unit 122 determines that the end condition is met ("Yes" in step S33), the process enters step S34. The end condition can be set according to the error, or it can be set according to the number of teacher images 43 processed, that is, the number of times of learning. Alternatively, the model generation unit 122 can use the provided verification data to evaluate the performance of the machine learning model, and end the machine learning when the evaluation meets the provided benchmark.

[0064] In step S34 , the model generation unit 122 outputs the machine learning model after machine learning as the learned model 20 . For example, the model generation unit 122 stores the learned model 20 in a predetermined storage device. The learned model 20 is used by the evaluation unit 13 .

[0065] (Evaluation related to coverage of objects)

[0066] refer to Figure 8 and Figure 9 , the evaluation related to the coverage of the object is explained. Figure 8 An example of such processing is shown in a flowchart as processing flow S4. Figure 9 1 is a diagram showing an example of image processing related to the processing flow S4. The processing flow S4 corresponds to the operation phase.

[0067] In step S41 , the coating determination unit 131 acquires an object image. For example, the coating determination unit 131 may acquire the object image 45 selected by the user or the object image 45 input from the pre-processing unit 11 .

[0068] In step S42, the coating identification unit 131 inputs the object image 45 into the learned model 20 and identifies the coating region of the object represented by the object image 45. The coating identification unit 131 obtains the estimation result based on the learned model 20 and identifies the coating region. For example, the coating identification unit 131 identifies a plurality of coating elements that form the coating region.

[0069] In step S43, the calculation unit 132 calculates an evaluation value related to the determined coating area. The calculation unit 132 can calculate an evaluation value related to the coverage of the substrate based on the determined coating area. As an example of an evaluation value related to coverage, a coverage ratio that indicates the proportion of the substrate covered by the coating area can be cited. The calculation unit 132 can calculate physical parameters such as area, height, radius, etc. for each of at least one coating element as an evaluation value. The height of a coating element refers to the distance from the substrate surface to the top of the coating element. The calculation unit 132 can calculate the height of a coating element located at the edge of an object based on the position of the substrate surface and the position of the top of the coating element. The calculation unit 132 can calculate an evaluation value related to shape, such as roundness, for each of at least one coating element. The calculation unit 132 can calculate statistical values related to multiple coating elements, such as standard deviation, mean value, and median value, as evaluation values. For example, the calculation unit 132 can calculate statistical values for various physical parameters such as area, roundness, height, and radius. When the coating element is a particle, the calculation unit 132 can calculate the coefficient of variation (CV) of the particle size as the evaluation value. The CV value is obtained by the following formula. The CV value is also an example of a statistical value.

[0070] CV value = (standard deviation) / (median diameter)

[0071] As an example of calculating the evaluation value, refer to Figure 9 , the calculation of coverage and related image processing are explained. Figure 9 The image 250 represents an object and two auxiliary images 260 and 270. The object image 250 represents a particulate material as an object, and the individual particles that are coating elements are represented by color or pattern. The auxiliary image 260 only represents the coating area, that is, multiple particles. The auxiliary image 270 shows the method for calculating the coverage. The auxiliary image 270 includes a virtual circle 271 representing the central area of the object described later and a virtual circle 272 representing the outer edge of the substrate. Within the virtual circle 271, only multiple particles are represented, as in the auxiliary image 260. Outside the virtual circle 271, the particulate material is represented in a form in which each particle is identified, as in the object image 250. The virtual circle 272 can be omitted.

[0072] In one example, the calculation unit 132 sets the central area of the object to calculate the coverage. The central area is a portion of the object represented by the image. For example, the calculation unit 132 sets the area from the center of the object to a predetermined radius as the central area. The calculation unit 132 calculates the evaluation value in the central area. For example, the calculation unit 132 calculates the total area of the coating area located in the central area, that is, the total area of one or more coating elements located in the central area. Then, the calculation unit 132 calculates the ratio of this total area to the area of the central area as the coverage. Each area can be determined by the number of pixels. Therefore, the calculation unit 132 can calculate the ratio of the total number of pixels in the coating area located in the central area to the number of pixels in the central area as the coverage. The calculation unit 132 can calculate the area of each coating element located in the central area as the evaluation value, or can calculate the evaluation value related to the shape. The calculation unit 132 can calculate a statistical value related to multiple coating elements located in the central area as the evaluation value. When the coating element is a particle, the calculation unit 132 may calculate the CV value in the central region as the evaluation value.

[0073] In step S44, the evaluation unit 13 outputs the evaluation value. The evaluation unit 13 may display the evaluation value on a monitor, store it in a predetermined storage device such as a database, or transmit it to another computer.

[0074] The evaluation unit 13 may display at least one of the object image 250 and the auxiliary images 260 and 270. That is, the evaluation unit 13 may display an image showing at least the determined coating area. The user can confirm the determination of the coating area and the basis of the evaluation value through this image.

[0075] The evaluation system 10 can execute the process S4 multiple times or repeatedly. For example, each time the user selects the target image 45, the evaluation system 10 executes the process S4 in response to the selection. The evaluation unit 13 can calculate a statistical value of multiple evaluation values corresponding to multiple objects as a further evaluation value.

[0076] [Modification]

[0077] The technology involved in the present invention has been described in detail above based on various examples. However, the present invention is not limited to the above embodiments. Various modifications can be made to the technology involved in the present invention without departing from its purpose.

[0078] The evaluation system does not need to include either the learning unit or the evaluation unit. The learned model can be ported between computer systems. Therefore, the learning unit can provide the learned model to other computer systems, and the evaluation unit can use the learned model provided by other computers.

[0079] The original image may be provided from a device other than the database, for example, it may be directly provided from an imaging device such as a SEM or a camera.

[0080] The preprocessing unit can be used only in either the learning phase or the application phase. Alternatively, the evaluation system may not include a preprocessing unit. The learning unit can obtain sample images from another computer system, and the evaluation unit can obtain object images from another computer system. The training image can be generated by another computer system, so the evaluation system does not need to include a training image generation unit.

[0081] At least one of the sample image, the teacher image, and the object image may be an image showing multiple objects. Therefore, the learning unit may generate a learned model based on a sample image showing multiple objects or a teacher image showing multiple objects. The learned model is a computational model learned by determining a coating area for at least one of the one or more objects shown in the object image. The evaluation unit may input an object image showing one or more objects into the learned model, and determine a coating area for at least one of the one or more objects. In this example, the evaluation unit calculates an evaluation value related to the determined coating area for at least one object.

[0082] The processing steps of the method executed by at least one processor are not limited to the examples in the above embodiments. For example, some of the above steps may be omitted, or the steps may be performed in another order. Furthermore, any two or more of the above steps may be combined, or some of the steps may be modified or deleted. Alternatively, other steps may be performed in addition to the above steps.

[0083] In the present invention, when comparing the magnitude relationship of two numerical values, either of the two criteria of “above” and “greater than” or “below” and “less than” may be used.

[0084] In the present invention, the expression "at least one processor executes a first process, executes a second process, ... executes an nth process" or a corresponding expression encompasses the case where the execution subject (i.e., the processor) of n processes from the first process to the nth process is changed midway. Specifically, this expression encompasses both the case where all n processes are executed by the same processor and the case where the processor is changed among the n processes using an arbitrary strategy.

[0085] [Note]

[0086] As is clear from the various examples described above, the present invention includes the following aspects.

[0087] (Note 1)

[0088] An evaluation system comprising at least one processor,

[0089] The at least one processor performs the following processing:

[0090] acquiring an object image representing an object having a substrate and a coating region on the substrate;

[0091] inputting the object image into a learned model for estimating the coating area based on the input image to determine the coating area of the object; and

[0092] An evaluation value related to the determined coating area is calculated.

[0093] (Note 2)

[0094] The evaluation system according to Supplementary Note 1, wherein:

[0095] The at least one processor calculates the evaluation value related to coverage of the substrate based on the determined coating area.

[0096] (Note 3)

[0097] The evaluation system according to Supplementary Note 2, wherein:

[0098] The at least one processor calculates a coverage ratio indicating a proportion of the substrate covered by the determined coating area as the evaluation value.

[0099] (Note 4)

[0100] The evaluation system according to Supplementary Note 1, wherein:

[0101] The at least one processor performs the following processing:

[0102] The object image is input into the learned model to determine a plurality of coating elements forming the coating area.

[0103] Statistical values related to the plurality of coating elements are calculated as the evaluation values.

[0104] (Note 5)

[0105] The evaluation system according to any one of Supplementary Notes 1 to 4, wherein:

[0106] The at least one processor performs the following processing:

[0107] setting a central area of the object represented by the object image; and

[0108] The evaluation value in the central area is calculated.

[0109] (Note 6)

[0110] The evaluation system according to Supplementary Note 5, wherein:

[0111] The at least one processor sets an area within a predetermined radius from the center of the object as the central area.

[0112] (Note 7)

[0113] The evaluation system according to any one of Supplementary Notes 1 to 6, wherein:

[0114] The at least one processor displays an image representing at least the determined coating area.

[0115] (Note 8)

[0116] The evaluation system according to any one of Supplementary Notes 1 to 7, wherein:

[0117] The at least one processor performs the following processing:

[0118] acquiring an original image representing at least one of the objects;

[0119] Converting the original image into a binary image;

[0120] performing distance transformation on the binary image, and determining the center and size of each of the at least one object; and

[0121] The object image representing the object entirely reflected in the original image is generated based on the center and the size of each of the at least one object.

[0122] (Note 9)

[0123] The evaluation system according to any one of Supplementary Notes 1 to 8, wherein:

[0124] The object is a particulate matter.

[0125] (Note 10)

[0126] The evaluation system according to Supplementary Note 9, wherein:

[0127] The substrate is a core particle,

[0128] The coating region is a collection of multiple particles.

[0129] (Note 11)

[0130] An evaluation method is performed by an evaluation system having at least one processor, the evaluation method comprising the following steps:

[0131] acquiring an object image representing an object having a substrate and a coating region on the substrate;

[0132] inputting the object image into a learned model for estimating the coating area based on the input image to determine the coating area of the object; and

[0133] An evaluation value related to the determined coating area is calculated.

[0134] (Note 12)

[0135] An evaluation program causes a computer to execute the following steps:

[0136] acquiring an object image representing an object having a substrate and a coating region on the substrate;

[0137] inputting the object image into a learned model for estimating the coating area based on the input image to determine the coating area of the object; and

[0138] An evaluation value related to the determined coating area is calculated.

[0139] According to Supplementary Notes 1, 11, and 12, the coating area of an object is estimated from the object image using the learned model. This structure makes it easy to identify the coating area and calculate the evaluation value related to the coating area. Therefore, it is easy to evaluate the coverage of the object.

[0140] According to Supplementary Note 2, the evaluation value related to the coverage of the base material by the coating area can be easily calculated.

[0141] According to Supplementary Note 3, the coverage based on the coating area can be easily calculated.

[0142] According to Supplementary Note 4, a plurality of coating elements can be easily identified using the learned model, and calculation of statistical values related to the coating elements also becomes easy.

[0143] According to Supplementary Note 5, an evaluation value is calculated for the central area of the object. Due to the object's three-dimensional shape, the object appears in the object image with a coating area concentrated around its periphery. Therefore, accurately calculating the evaluation value for the coating area can be difficult. In contrast, the central area of the object appears in the object image as if it were actually in the state. Therefore, calculating the evaluation value for this central area allows for more accurate evaluation of the object's coverage.

[0144] According to Supplementary Note 6, a circular central area can be set, and therefore the central area can be appropriately set regardless of the outer shape of the object.

[0145] According to Supplementary Note 7, the specified coating area can be displayed, so the processing results based on the learned model can be presented to the user. The user can refer to the results and gain a sense of recognition of the evaluation related to the coating of the object.

[0146] According to Supplementary Note 8, the center and size of each object are determined by converting the image to a binary image and performing distance conversion. Based on this determination, an object image representing the object as a whole within the original image can be obtained. This method automatically allows for the acquisition of an appropriate object image for evaluating object coverage.

[0147] According to Supplementary Note 9, evaluation regarding the coverage with the particulate matter can be easily performed.

[0148] According to Supplementary Note 10, the learned model makes it easy to identify a set of multiple particles and to calculate an evaluation value related to the set. Therefore, it is easy to evaluate the coverage of particulate matter including particles as components of the coating area.

[0149] Explanation of symbols

[0150] 10-Evaluation system, 11-Preprocessing unit, 12-Learning unit, 13-Evaluation unit, 20-Learning completed model, 31-1st original image database, 32-Teacher image database, 33-2nd original image database, 41-1st original image, 42-Sample image, 43-Teacher image, 44-2nd original image, 45-Object image, 121-Teacher image generation unit, 122-Model generation unit, 131-Coating determination unit, 132-Calculation unit.

Claims

1. An evaluation system comprising at least one processor, The at least one processor performs the following processing: acquiring an object image representing an object having a substrate and a coating region on the substrate; inputting the object image into a learned model for estimating the coating area based on the input image to determine the coating area of the object; and An evaluation value related to the determined coating area is calculated.

2. The evaluation system according to claim 1, wherein: The at least one processor calculates the evaluation value related to coverage of the substrate based on the determined coating area.

3. The evaluation system according to claim 2, wherein: The at least one processor calculates a coverage ratio indicating a proportion of the substrate covered by the determined coating area as the evaluation value.

4. The evaluation system according to claim 1, wherein: The at least one processor performs the following processing: The object image is input into the learned model to determine a plurality of coating elements forming the coating area. Statistical values related to the plurality of coating elements are calculated as the evaluation values.

5. The evaluation system according to any one of claims 1 to 4, wherein: The at least one processor performs the following processing: setting a central area of the object represented by the object image; and The evaluation value in the central area is calculated.

6. The evaluation system according to claim 5, wherein: The at least one processor sets an area within a predetermined radius from the center of the object as the central area.

7. The evaluation system according to any one of claims 1 to 4, wherein: The at least one processor displays an image representing at least the determined coating area.

8. The evaluation system according to any one of claims 1 to 4, wherein: The at least one processor performs the following processing: acquiring an original image representing at least one of the objects; Converting the original image into a binary image; performing distance transformation on the binary image, and determining the center and size of each of the at least one object; and The object image representing the object entirely reflected in the original image is generated based on the center and the size of each of the at least one object.

9. The evaluation system according to any one of claims 1 to 4, wherein: The object is a particulate matter.

10. The evaluation system according to claim 9, wherein: The substrate is a core particle, The coating region is a collection of multiple particles.

11. An evaluation method, performed by an evaluation system having at least one processor, comprising the following steps: acquiring an object image representing an object having a substrate and a coating region on the substrate; inputting the object image into a learned model for estimating the coating area based on the input image to determine the coating area of the object; and An evaluation value related to the determined coating area is calculated.

12. An evaluation program causing a computer to execute the following steps: acquiring an object image representing an object having a substrate and a coating region on the substrate; inputting the object image into a learned model for estimating the coating area based on the input image to determine the coating area of the object; and An evaluation value related to the determined coating area is calculated.

Images