JP2020201051A - Evaluation device and evaluation system - Google Patents

Abstract

To provide an evaluation device and evaluation system capable of improving evaluation accuracy.SOLUTION: An evaluation device 1 for evaluating three-dimensional data based on a front surface state of an evaluation object 3 includes: an acquisition section for acquiring three-dimensional data having multiple pieces of data, i.e., position data indicating a coordinate of a three-dimensional space with respect to the evaluation object 3 and color data indicating color in the position data of the evaluation object; a reference database for storing a learning model having association between previously acquired past three-dimensional data and reference information associated with the past three-dimensional data; and an evaluation section for generating an evaluation result with respect to three-dimensional data with referring to the reference database.SELECTED DRAWING: Figure 1

Description

The present invention relates to an evaluation device and an evaluation system for evaluating three-dimensional data based on the surface state of an evaluation object.

Conventionally, as a method for evaluating an evaluation object such as a three-dimensional object, for example, a three-dimensional object detection device of Patent Document 1 has been proposed.

The 3D object detection device disclosed in Patent Document 1 calculates the posture of a 3D model corresponding to a 3D object by using a distance image acquisition unit (camera) that acquires a distance image of the 3D object and the distance image. Predetermined from the 3D model rotating part, the projected image edge extraction part that extracts the edge of the projected image obtained by projecting the 3D model on a plane, and the point group at the position closest to the camera in the extracted edge. Matching process that detects the posture of a 3D object by extracting the points of the 3D model within the distance as the matching points and matching the extracted matching points with the matching points obtained from the distance image. It has a part.

Japanese Unexamined Patent Publication No. 2019-60695

Here, it is required to improve the evaluation accuracy of the method of evaluating the surface state of the evaluation object. In this regard, Patent Document 1 only discloses a technique capable of detecting the actual posture of a three-dimensional object even if there is an error between the actual posture of the three-dimensional object and the estimated posture. The evaluation accuracy of the surface condition cannot be improved.

Therefore, the present invention has been devised in view of the above-mentioned problems, and an object of the present invention is to provide an evaluation device and an evaluation system capable of improving evaluation accuracy.

The evaluation device according to the first invention is an evaluation device that evaluates three-dimensional data based on the surface state of the evaluation target, and is position data indicating coordinates in three-dimensional space with respect to the evaluation target, and the evaluation target. The association between the acquisition unit that acquires the three-dimensional data having a plurality of color data indicating the colors in the position data, the past three-dimensional data acquired in advance, and the reference information associated with the past three-dimensional data. It is characterized by including a reference database in which a learning model having a property is stored, and an evaluation unit that refers to the reference database and generates an evaluation result for the three-dimensional data.

The evaluation device according to the second invention is characterized in that, in the first invention, the three-dimensional data includes the position data and the time data indicating the time when the color data is imaged.

In the evaluation device according to the third invention, in the first invention or the second invention, the three-dimensional data has the position data and the color data imaged from two or more different directions with respect to the evaluation object. It is characterized by.

In the third invention, the evaluation device according to the fourth invention has the position data and the color data imaged from three directions 120 degrees different from the evaluation target at the same distance. It is a feature.

In any of the first to fourth inventions, the evaluation device according to the fifth invention includes a sampling data generation unit that generates a plurality of sampling data based on the plurality of position data, and a plurality of the evaluation units. A group data generation unit that generates a plurality of group data classified by the same or similar data among the sampling data, the plurality of the position data, and the plurality of the color data, and the reference database are referred to, and the plurality of the above It is characterized by having an evaluation result generation unit that generates the evaluation result based on group data.

The evaluation device according to the sixth invention is constructed by machine learning using the past three-dimensional data and the reference information as learning data in any one of the first to fifth inventions. It is characterized by that.

The evaluation system according to the seventh invention is an evaluation system that evaluates three-dimensional data based on the surface state of the evaluation object, and is position data indicating coordinates in three-dimensional space with respect to the evaluation object, and the evaluation object. The association between the acquisition means for acquiring the three-dimensional data having a plurality of color data indicating the colors in the position data, the past three-dimensional data acquired in advance, and the reference information associated with the past three-dimensional data. It is characterized by comprising a reference database in which a learning model having a property is stored, and an evaluation means for referring to the reference database and generating an evaluation result for the three-dimensional data.

According to the first to sixth inventions, the evaluation unit refers to the reference database and generates the evaluation result for the three-dimensional data. Therefore, the evaluation result based on the past result can be generated, and the color data can be the evaluation target in addition to the position data with respect to the evaluation target. This makes it possible to improve the evaluation accuracy.

In particular, according to the second invention, the three-dimensional data has time data. Therefore, even when the surface state of the evaluation target changes with time, it is possible to carry out the evaluation based on the change with time. This makes it possible to expand the evaluation conditions.

In particular, according to the third invention, the three-dimensional data has position data and color data imaged from two or more different directions with respect to the evaluation target. Therefore, the accuracy of the relative positional relationship for each position data can be improved. This makes it possible to further improve the evaluation accuracy.

In particular, according to the fourth invention, the three-dimensional data has position data and color data imaged from three directions that differ by 120 degrees at the same distance with respect to the evaluation object. Therefore, it is possible to evaluate the three-dimensional data based on the surface state of the entire circumference of the evaluation object. This makes it possible to further improve the evaluation accuracy.

In particular, according to the fifth invention, the evaluation unit has a group data generation unit and an evaluation result generation unit. Therefore, the evaluation result can be generated in the state where the position data and the color data are linked. This makes it possible to evaluate the difference in color scheme of the evaluation target. Therefore, the types of evaluation objects that can be evaluated can be dramatically increased.

In particular, according to the sixth invention, the learning model is constructed by machine learning using past three-dimensional data and reference information as learning data. Therefore, even when evaluating unknown three-dimensional data different from the past three-dimensional data, the quantitative evaluation can be performed. This makes it possible to further improve the evaluation accuracy.

According to the seventh invention, the evaluation means refers to the reference database and generates the evaluation result for the three-dimensional data. Therefore, the evaluation result based on the past result can be generated, and the color data can be the evaluation target in addition to the position data with respect to the evaluation target. This makes it possible to improve the evaluation accuracy.

FIG. 1 is a schematic diagram showing an example of an evaluation system according to the present embodiment. FIG. 2 is a schematic diagram showing an example of the operation of the evaluation system. 3 (a) to 3 (c) are schematic views showing an example of an output screen in the evaluation system. FIG. 4A is a schematic diagram showing an example of the configuration of the evaluation device according to the present embodiment, and FIG. 4B is a schematic diagram showing an example of the function of the evaluation device according to the present embodiment. FIG. 5 is a schematic diagram showing an example of a reference database according to the present embodiment. FIG. 6 is a schematic view showing a first modification of the reference database in the present embodiment. FIG. 7 is a schematic view showing a second modification of the reference database in the present embodiment. FIG. 8 is a flowchart showing an example of the operation of the evaluation system in the present embodiment. FIG. 9 is a schematic diagram showing an example of the evaluation means in the present embodiment.

Hereinafter, an example of the evaluation system and the evaluation device according to the embodiment to which the present invention is applied will be described with reference to the drawings.

An example of the evaluation system 100 and the evaluation device 1 in the present embodiment will be described with reference to FIG.

The evaluation system 100 in the present embodiment has an evaluation device 1 as shown in FIG. 1, for example. The evaluation device 1 may be connected to, for example, an image pickup device 2, or may be connected to another terminal 5 or a server 6 via, for example, a communication network 4.

The evaluation system 100 evaluates three-dimensional data based on the surface state of the evaluation object 3. The evaluation system 100 is used, for example, as an inspection machine when manufacturing the evaluation object 3, is also used for evaluation of similar products, and can evaluate a product having a three-dimensional structure as the evaluation object 3. The "surface state" indicates the shape and color scheme of the evaluation object 3, and also indicates defects such as cracks, stains, color unevenness, and adhesion of foreign matter.

As shown in FIG. 2, for example, the evaluation system 100 refers to a reference database described later after the evaluation device 1 acquires the three-dimensional data, and generates an evaluation result for the three-dimensional data. The evaluation system 100 outputs, for example, output information corresponding to the evaluation result (output information “rabbit” is output in FIG. 2). The three-dimensional data has a plurality of position data and color data associated with the position data.

The position data shows the coordinates in the three-dimensional space with respect to the evaluation object 3, for example, the parameters "x, y, z" in the Cartesian coordinate system, the parameters "r, θ, φ" in the spherical coordinate system, and the like. The color data indicates the color for each position data of the evaluation object 3, for example, the parameters “r, g, b” of the RGB color model. That is, by using a plurality of position data and color data included in the three-dimensional data, the surface state of the evaluation object 3 can be reproduced. For example, the three-dimensional data is color data for a plurality of position data. It is indicated by a point cloud (point cloud) expressed in the color of.

The three-dimensional data may include, for example, time data. The time data indicates the time when the position data and the color data were captured. In this case, even when the surface state of the evaluation object 3 changes with time, it is possible to carry out the evaluation based on before and after the change with time.

The three-dimensional data is generated by, for example, imaging the evaluation object 3 using the imaging device 2, and for example, a plurality of three-dimensional data generated by imaging one evaluation object 3 using a plurality of imaging devices 2. The evaluation device 1 may acquire the data as one three-dimensional data. The image pickup device 2 may be built in, for example, the evaluation device 1.

The three-dimensional data includes, for example, position data and color data captured from two or more different directions with respect to the evaluation object 3. In this case, the accuracy of the relative positional relationship for each position data can be improved.

The three-dimensional data includes, for example, position data and color data taken from three directions at the same distance and different arbitrary angles (for example, θ1, θ2, θ3) with respect to the evaluation object 3. In particular, when the angles θ1, θ2, and θ3 are set, three-dimensional data based on the surface state of the entire circumference of the evaluation object 3 can be generated while the number of image pickup devices 2 is suppressed.

The 3D data is generated using a known 3D camera (3D camera), for example using Real Sense®. The three-dimensional data may include, for example, position data generated by using a known distance image camera and color data generated by using a known RGB camera.

As shown in FIG. 3A, for example, the evaluation device 1 may display an image that reproduces the surface state of the evaluation object 3 by using a plurality of position data based on the acquired three-dimensional data. Further, as shown in FIG. 3B, for example, the evaluation device 1 displays an image that reproduces the surface state of the evaluation object 3 by using a plurality of position data and color data based on the acquired three-dimensional data. You may. As shown in FIG. 3C, for example, the evaluation device 1 has details of the “detection result” (for example, “normal” or “abnormal”) and the “type” of the evaluation object 3 as output information corresponding to the evaluation result. "(For example," rabbit ") may be displayed.

(Evaluation device 1)
Next, an example of the evaluation device 1 in the present embodiment will be described with reference to FIG. FIG. 4A is a schematic diagram showing an example of the configuration of the evaluation device 1 in the present embodiment, and FIG. 4B is a schematic diagram showing an example of the function of the evaluation device 1 in the present embodiment.

As the evaluation device 1, for example, an electronic device such as a personal computer (PC) is used, and for example, an electronic device such as a smartphone, a tablet terminal, a wearable terminal, an IoT (Internet of Things) device, a Raspberry Pi (registered trademark), etc. A single board computer may be used. As shown in FIG. 4A, for example, the evaluation device 1 includes a housing 10, a CPU (Central Processing Unit) 101, a ROM (Read Only Memory) 102, a RAM (Random Access Memory) 103, and a storage unit. It includes 104 and I / F 105-107. Each configuration 101-107 is connected by an internal bus 110.

The CPU 101 controls the entire evaluation device 1. The ROM 102 stores the operation code of the CPU 101. The RAM 103 is a work area used during the operation of the CPU 101. The storage unit 104 stores various information such as three-dimensional data and a reference database. As the storage unit 104, for example, in addition to an HDD (Hard Disk Drive), a data storage device such as an SSD (solid state drive) or a floppy disk is used. For example, the evaluation device 1 may have a GPU (Graphics Processing Unit) (not shown). Having a GPU enables faster arithmetic processing than usual.

The I / F 105 is an interface for transmitting and receiving various information with the image pickup device 2, and also transmits and receives various information with other terminals 5 and servers 6 via a communication network 4 such as the Internet. It may be an interface for. The I / F 106 is an interface for transmitting / receiving information to / from the input portion 108. For example, a keyboard is used as the input portion 108, and the administrator of the evaluation device 1 inputs various information or control commands of the evaluation device 1 via the input portion 108. The I / F 107 is an interface for transmitting and receiving various information to and from the output portion 109. The output unit 109 outputs various information stored in the storage unit 104, the processing status of the evaluation device 1, and the like. A display is used as the output portion 109, and for example, a touch panel type may be used.

<Reference database>
The learning model is stored in the reference database stored in the storage unit 104. The learning model has a relationship between the past three-dimensional data acquired in advance and the reference information associated with the past three-dimensional data. For example, past three-dimensional data and reference information may be stored in the reference database. The learning model is constructed by machine learning using a plurality of learning data, for example, using past three-dimensional data and reference information as a set of learning data. As a learning method, deep learning such as a convolutional neural network is used.

In this case, for example, association indicates the degree of connection between many-to-many information. The associations are updated as appropriate during the machine learning process. For this reason, the learning model presents, for example, a classifier with associations (eg, functions) optimized based on past 3D data and reference information. Therefore, the evaluation result for the three-dimensional data is generated by using the learning model constructed based on all the results of evaluating the surface state of the evaluation object 3 in the past. As a result, even when the evaluation object 3 has a complicated surface state, the optimum evaluation result can be generated. Further, the optimum evaluation result can be quantitatively generated not only when the three-dimensional data is the same as or similar to the past three-dimensional data but also when it is dissimilar. By increasing the generalization ability when performing machine learning, it is possible to improve the evaluation accuracy for unknown three-dimensional data.

In addition, the association may have a plurality of association degrees indicating the degree of connection between each past three-dimensional data and each reference information, for example. The degree of association can correspond to a weight variable, for example, when the learning model is constructed by a neural network.

The past three-dimensional data shows the same kind of information as the above-mentioned three-dimensional data. The past three-dimensional data includes, for example, a plurality of three-dimensional data acquired when the evaluation object 3 was evaluated in the past.

The reference information is linked to the past three-dimensional data and indicates information on the surface state of the evaluation object 3. The reference information includes, for example, type information indicating the type of the evaluation object 3 (for example, "rabbit" or "cup ramen"), or the production standard of the evaluation object 3 (for example, "within standard", "non-standard", "foreign substance". Includes standard information such as "Yes", "With color unevenness", "With cracks", etc.). The specific contents included in the reference information can be arbitrarily set.

As shown in FIG. 5, for example, the association may indicate the degree of connection between the past three-dimensional data and the reference information. In this case, by using the association, a plurality of data included in the reference information (FIG. 5) for each of the plurality of data (“data A” to “data C” in FIG. 5) included in the past three-dimensional data. In 5, the degree of relationship between "reference A" to "reference C") can be associated and stored. Therefore, for example, it is possible to link a plurality of data included in the reference information to one data included in the past three-dimensional data through association, and it is possible to generate a multifaceted evaluation result. be able to.

The association has, for example, a plurality of association degrees for associating a plurality of data included in the past three-dimensional data with a plurality of data included in the reference information. The degree of association is indicated by three or more stages such as percentage, 10 stages, or 5 stages, and is indicated by, for example, line characteristics (for example, thickness). For example, the "data A" included in the past three-dimensional data indicates the degree of association AA "75%" with the "reference A" included in the reference information, and the "reference B" included in the reference information. The degree of association between them is AB "12%". That is, the "degree of association" indicates the degree of connection between each data. For example, the higher the degree of association, the stronger the connection of each data. When constructing a learning model by the above-mentioned machine learning, the association may be set to have three or more levels of association.

As shown in FIG. 6, for example, the past three-dimensional data may be stored in the reference database by dividing the past position data and the past color data. In this case, the degree of association is calculated based on the relationship between the combination of the past position data and the past color data and the reference information. In addition to the above, the past three-dimensional data may be stored in the reference database by dividing at least one of the past three-dimensional data and the past color data.

For example, the combination of "position A" included in the past position data and "color A" included in the past color data indicates the degree of association AAA "63%" with "reference A", and "reference". The degree of association with "B" is ABA "63%". In this case, the past position data and the past color data can be stored independently. Therefore, when generating the evaluation result, it is possible to improve the accuracy and expand the range of options.

The association may include synthetic data and similarity, as shown in FIG. 7, for example. The composite data is indicated by three or more levels of similarity with past position data or past color data. The composite data is stored in the reference database in the form of a numerical value, a matrix, a histogram, or the like, or may be stored in a format such as an image or a character string.

FIG. 4B is a schematic diagram showing an example of the function of the evaluation device 1. The evaluation device 1 includes an acquisition unit 11, a storage unit 12, an evaluation unit 13, and an output unit 14, and may include, for example, an update unit 15. Each function shown in FIG. 4B is realized by the CPU 101 executing a program stored in the storage unit 104 or the like using the RAM 103 as a work area, and may be controlled by, for example, artificial intelligence.

<Acquisition unit 11>
The acquisition unit 11 acquires three-dimensional data. The acquisition unit 11 acquires three-dimensional data from the imaging device 2 and the like, and also acquires three-dimensional data from, for example, a built-in imaging unit (not shown). The frequency and cycle for the acquisition unit 11 to acquire the three-dimensional data are arbitrary.

The acquisition unit 11 receives various information transmitted to the evaluation device 1. The acquisition unit 11 may receive various information such as three-dimensional data transmitted from an external terminal such as another terminal 5 via, for example, the communication network 4 and the I / F 105.

<Memory unit 12>
The storage unit 12 retrieves various information such as a reference database stored in the storage unit 104 as needed. The storage unit 12 stores various information acquired or generated by the configurations 11 and 13 to 15 in the storage unit 104.

<Evaluation unit 13>
The evaluation unit 13 refers to the reference database and generates an evaluation result for the three-dimensional data. The evaluation unit 13 uses, for example, three-dimensional data as input data, selects the optimum reference information associated with the solution calculated based on the learning model, and generates an evaluation result based on the optimum reference information.

When referring to the reference database shown in FIG. 5, for example, the evaluation unit 13 selects data that is the same as or similar to the data included in the three-dimensional data (for example, “data A”: the first data). As the first data, data that partially matches or completely matches the three-dimensional data is selected, and for example, similar data is selected. When the three-dimensional data is represented by a numerical value such as a matrix, the numerical range included in the selected first data may be set in advance.

The evaluation unit 13 selects the reference information associated with the selected first data and the degree of association (first degree of association) between the selected first data and the reference information, and selects the selected reference information and the first degree of association. The evaluation result is generated based on. The first degree of association is selected from the associations constructed in advance, and may be calculated by the evaluation unit 13.

For example, the evaluation unit 13 has the data "reference A" included in the reference information associated with the first data "data A", and the first degree of association (linkage degree AA) between the "data A" and the "reference A". Select "75%". The reference information and the first degree of association may include a plurality of data. In this case, in addition to the above-mentioned "reference A" and "75%", the reference information "reference B" associated with the first data "data A", and the second between "data A" and "reference B". 1 The degree of association (degree of association AB) "12%" may be selected and the evaluation result may be generated based on "reference A" and "75%", and "reference B" and "12%".

The evaluation result may include three-dimensional data. Further, the first degree of association is indicated by three or more stages such as a percentage.

The evaluation unit 13 uses format data such as an output format stored in advance in the storage unit 104 or the like to provide a format (for example, a character string) that the user can understand the above-selected reference information and the first association degree. Generate the evaluation result shown. For example, a known technique may be used for setting the format when generating the evaluation result.

The evaluation unit 13 determines the content of the evaluation result based on, for example, the selected first degree of association. For example, the evaluation unit 13 generates an evaluation result based on the reference information associated with the first association degree of "50%" or more, and uses the reference information associated with the first association degree of less than "50%" as the evaluation result. It may be set not to reflect. As the determination standard based on the first degree of association, for example, the administrator or the like may set a threshold value or the like in advance, and the threshold range or the like can be arbitrarily set. Further, the evaluation unit 13 may determine the content of the evaluation result based on, for example, the result of calculating two or more first association degrees and the comparison of two or more first association degrees.

The evaluation unit 13 may have a sampling data generation unit 13a, a group data generation unit 13b, and an evaluation result generation unit 13c, as shown in FIG. 9, for example.

<< Sampling data generation unit 13a >>
The sampling data generation unit 13a generates a plurality of sampling data based on the plurality of position data. The sampling data generation unit 13a generates sampling data from the position data of the points selected by the preset algorithm. As the algorithm, for example, a known technique (for example, the Farstest Point Sampling algorithm) is used.

<< Group data generation unit 13b >>
The group data generation unit 13b generates a plurality of group data classified by the same or similar data among the plurality of sampling data, the plurality of position data, and the plurality of color data. The group data generation unit 13b generates, for example, a plurality of position data and color data included within a certain distance from each sampling data as group data. In this case, for example, one group data includes one sampling data, a plurality of position data, and a plurality of color data.

<< Evaluation result generation unit 13c >>
The evaluation result generation unit 13c refers to the reference database and generates an evaluation result based on a plurality of group data. The evaluation result generation unit 13c performs, for example, a two-dimensional convolution process for each group data and a pooling process for each group data, and generates an evaluation result corresponding to the process result. The evaluation result generation unit 13c selects, for example, reference information corresponding to the processing result, and generates an evaluation result based on the reference information.

In addition, each generation unit 13a, 13b, 13c may execute the above-mentioned processing with reference to the learning model constructed by using the deep learning framework which can correspond to 3D data such as PointNet.

<Output unit 14>
The output unit 14 outputs the evaluation result. The output unit 14 transmits the evaluation result to the output unit 109 via the I / F 107, and also transmits the evaluation result to another terminal 5 or the like via, for example, the I / F 105. The output unit 14 outputs data for displaying an image that reproduces the surface state of the evaluation object 3 shown in FIG. 3, for example, to the output unit 109 or the like.

<Update part 15>
The update unit 15 updates, for example, the reference database. When the update unit 15 newly acquires the relationship between the past three-dimensional data and the reference information, the update unit 15 reflects the relationship in the association. For example, when the administrator or the like determines the accuracy of the content of the evaluation result based on the evaluation result generated by the evaluation unit 13 and the evaluation device 1 acquires the determination result, the update unit 15 puts it in the reference database based on the determination result. Update the included associations.

<Communication network 4>
The communication network 4 is, for example, an Internet network or the like to which the evaluation device 1 and the like are connected via a communication circuit. The communication network 4 may be composed of a so-called optical fiber communication network. Further, the communication network 4 may be realized by a known communication network such as a wireless communication network in addition to the wired communication network.

<Other terminals 5>
As the other terminal 5, for example, a terminal embodied in an electronic device similar to the evaluation device 1 is used. The other terminal 5 indicates, for example, a central control device capable of communicating with a plurality of evaluation devices 1.

<Server 6>
For example, the server 6 stores the above-mentioned various information. Various information sent via, for example, the communication network 4 is stored in the server 6. For example, information similar to that of the storage unit 104 may be stored in the server 6, and various information may be transmitted and received to and from the evaluation device 1 via the communication network 4. That is, the evaluation device 1 may use the server 6 instead of the storage unit 104.

(Example of operation of evaluation system 100)
Next, an example of the operation of the evaluation system 100 in this embodiment will be described. FIG. 8 is a flowchart showing an example of the operation of the evaluation system 100 in the present embodiment.

<Acquisition means S110>
As shown in FIG. 8, three-dimensional data is acquired (acquisition means S110). The acquisition unit 11 acquires three-dimensional data including position data and color data from, for example, the image pickup apparatus 2. The acquisition unit 11 stores the three-dimensional data in the storage unit 104, for example, via the storage unit 12.

For example, when one evaluation object 3 is imaged by using a plurality of image pickup devices 2, the acquisition unit 11 acquires a plurality of three-dimensional data generated by the plurality of image pickup devices 2 as one three-dimensional data. The acquisition unit 11 may acquire the three-dimensional data every time the three-dimensional data is generated, or may acquire the three-dimensional data generated in an arbitrary period, for example.

<Evaluation means S120>
Next, the reference database is referred to, and the evaluation result for the three-dimensional data is generated (evaluation means S120). The evaluation unit 13 acquires the three-dimensional data acquired by the acquisition unit 11, and acquires, for example, the reference database stored in the storage unit 104. The evaluation unit 13 uses, for example, three-dimensional data as input data, selects the optimum reference information associated with the solution calculated based on the association shown by the function or the like, and generates an evaluation result based on the optimum reference information. ..

The evaluation unit 13 may generate one evaluation result for one three-dimensional data, or may generate one evaluation result for, for example, a plurality of three-dimensional data. The evaluation unit 13 generates an evaluation result using, for example, format data such as an output format stored in the storage unit 104. The evaluation unit 13 stores the evaluation result in the storage unit 104, for example, via the storage unit 12.

In the evaluation means S120, for example, the sampling data generation unit 13a described above may generate a plurality of sampling data, the group data generation unit 13b may generate group data, and the evaluation result generation unit 13c may generate the evaluation result. ..

After that, for example, the evaluation result may be output (output means S130). The output unit 14 outputs the evaluation result to the output unit 109 or the like, and also outputs the evaluation result to another terminal 5 via, for example, the communication network 4.

<Update means S140>
In addition, for example, when the relationship between the past three-dimensional data and the reference information is newly acquired, the relationship may be reflected in the association (update means S140). For example, when the evaluation device 1 acquires a judgment result in which the administrator or the like judges the accuracy of the evaluation result based on the evaluation result generated by the evaluation unit 13, the update unit 15 includes the association included in the reference database based on the judgment result. Update sex.

As a result, the operation of the evaluation system 100 in the present embodiment may be completed. The timing when the update means S140 is implemented is arbitrary.

According to the present embodiment, the evaluation unit 13 refers to the reference database and generates an evaluation result for the three-dimensional data. Therefore, the evaluation result based on the past result can be generated, and the color data can be the evaluation target in addition to the position data with respect to the evaluation object 3. This makes it possible to improve the evaluation accuracy.

Further, according to the present embodiment, the three-dimensional data has time data. Therefore, even when the surface state of the evaluation object 3 changes with time, it is possible to carry out the evaluation based on the time before and after the change. This makes it possible to expand the evaluation conditions.

Further, according to the present embodiment, the three-dimensional data has position data and color data captured from two or more different directions with respect to the evaluation object 3. Therefore, the accuracy of the relative positional relationship for each position data can be improved. This makes it possible to further improve the evaluation accuracy.

Further, according to the present embodiment, the three-dimensional data has position data and color data imaged from three directions that differ by 120 degrees at the same distance with respect to the evaluation object 3. Therefore, it is possible to evaluate the three-dimensional data based on the surface state of the entire circumference of the evaluation object 3. This makes it possible to further improve the evaluation accuracy.

Further, according to the present embodiment, the evaluation unit 13 has a group data generation unit 13b and an evaluation result generation unit 13c. Therefore, the evaluation result can be generated in the state where the position data and the color data are linked. As a result, it is possible to evaluate the difference in the color scheme of the evaluation object 3. Therefore, the types of evaluation objects 3 that can be evaluated can be dramatically increased.

Further, according to the present embodiment, the learning model is constructed by machine learning using past three-dimensional data and reference information as learning data. Therefore, even when evaluating unknown three-dimensional data different from the past three-dimensional data, the quantitative evaluation can be performed. This makes it possible to further improve the evaluation accuracy.

Further, according to the present embodiment, the evaluation means S120 refers to the reference database and generates an evaluation result for the three-dimensional data. Therefore, the evaluation result based on the past result can be generated, and the color data can be the evaluation target in addition to the position data with respect to the evaluation object 3. This makes it possible to improve the evaluation accuracy.

Although embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other embodiments, and various omissions, replacements, and changes can be made without departing from the gist of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are also included in the scope of the invention described in the claims and the equivalent scope thereof.

1: Evaluation device 2: Image pickup device 3: Evaluation object 4: Communication network 5: Terminal 6: Server 10: Housing 11: Acquisition unit 12: Storage unit 13: Evaluation unit 13a: Sampling data generation unit 13b: Group data generation Unit 13c: Evaluation result generation unit 14: Output unit 15: Update unit 100: Evaluation system 101: CPU
102: ROM
103: RAM
104: Preservation unit 105: I / F
106: I / F
107: I / F
108: Input part 109: Output part 110: Internal bus S110: Acquisition means S120: Evaluation means S130: Output means S140: Update means

The evaluation device according to the first invention is an evaluation device that evaluates three-dimensional data based on the surface state of the evaluation target, and is position data indicating coordinates in three-dimensional space with respect to the evaluation target, and the evaluation target. The association between the acquisition unit that acquires the three-dimensional data having a plurality of color data indicating the colors in the position data, the past three-dimensional data acquired in advance, and the reference information associated with the past three-dimensional data. The evaluation unit includes a reference database in which a learning model having a property is stored, an evaluation unit that refers to the reference database and generates an evaluation result for the three-dimensional data, and the evaluation unit is based on a plurality of the position data. A sampling data generation unit that generates a plurality of sampling data, and a plurality of group data classified by the same or similar data among the plurality of the sampling data, the plurality of the position data, and the plurality of the color data are generated. and group data generation unit, with reference to the reference database, and wherein the evaluation result generation unit that generates the evaluation results based on a plurality of the group data, the Rukoto to have a.

The evaluation device according to the fifth invention is constructed by machine learning using the past three-dimensional data and the reference information as learning data in any one of the first to fourth inventions. It is characterized by that.

The evaluation system according to the sixth invention is an evaluation system that evaluates three-dimensional data based on the surface state of the evaluation object, and is position data indicating coordinates in the three-dimensional space with respect to the evaluation object, and the evaluation object. The association between the acquisition means for acquiring the three-dimensional data having a plurality of color data indicating the colors in the position data, the past three-dimensional data acquired in advance, and the reference information associated with the past three-dimensional data. A reference database in which a learning model having a property is stored and an evaluation means for referring to the reference database and generating an evaluation result for the three-dimensional data are provided , and the evaluation means is based on a plurality of the position data. A sampling data generation means for generating a plurality of sampling data, and a plurality of group data classified by the same or similar data among the plurality of the sampling data, the plurality of the position data, and the plurality of the color data are generated. and group data generating means, by referring to the reference database, and wherein the evaluation result generating means for generating the evaluation results based on a plurality of the group data, the Rukoto to have a.

According to the first to fifth inventions, the evaluation unit refers to the reference database and generates the evaluation result for the three-dimensional data. Therefore, the evaluation result based on the past result can be generated, and the color data can be the evaluation target in addition to the position data with respect to the evaluation target. This makes it possible to improve the evaluation accuracy.

According to the first to fifth inventions, the evaluation unit includes a group data generation unit and an evaluation result generation unit. Therefore, the evaluation result can be generated in the state where the position data and the color data are linked. This makes it possible to evaluate the difference in color scheme of the evaluation target. Therefore, the types of evaluation objects that can be evaluated can be dramatically increased.

In particular, according to the fifth invention, the learning model is constructed by machine learning using past three-dimensional data and reference information as learning data. Therefore, even when evaluating unknown three-dimensional data different from the past three-dimensional data, the quantitative evaluation can be performed. This makes it possible to further improve the evaluation accuracy.

According to the sixth invention, the evaluation means refers to the reference database and generates the evaluation result for the three-dimensional data. Therefore, the evaluation result based on the past result can be generated, and the color data can be the evaluation target in addition to the position data with respect to the evaluation target. This makes it possible to improve the evaluation accuracy.
Further, according to the sixth invention, the evaluation means includes a group data generation means and an evaluation result generation means. Therefore, the evaluation result can be generated in the state where the position data and the color data are linked. This makes it possible to evaluate the difference in color scheme of the evaluation target. Therefore, the types of evaluation objects that can be evaluated can be dramatically increased.

Claims (7)

An evaluation device that evaluates three-dimensional data based on the surface condition of the evaluation object.
An acquisition unit that acquires the three-dimensional data having a plurality of position data indicating the coordinates of the three-dimensional space with respect to the evaluation object and color data indicating the color in the position data of the evaluation object.
A reference database in which a learning model having a relationship between the past 3D data acquired in advance and the reference information associated with the past 3D data is stored.
An evaluation unit that refers to the reference database and generates evaluation results for the three-dimensional data,
An evaluation device characterized by comprising. The evaluation device according to claim 1, wherein the three-dimensional data includes the position data and the time data indicating the time when the color data is captured. The evaluation device according to claim 1 or 2, wherein the three-dimensional data has the position data and the color data imaged from two or more different directions with respect to the evaluation object. The evaluation device according to claim 3, wherein the three-dimensional data has the position data and the color data imaged from three directions that differ by 120 degrees at the same distance with respect to the evaluation object. The evaluation unit
A sampling data generator that generates a plurality of sampling data based on the plurality of position data,
A group data generation unit that generates a plurality of group data classified by the same or similar data among the plurality of sampling data, the plurality of position data, and the plurality of color data.
An evaluation result generation unit that refers to the reference database and generates the evaluation result based on a plurality of the group data.
The evaluation device according to any one of claims 1 to 4, wherein the evaluation device is characterized by having. The evaluation device according to any one of claims 1 to 5, wherein the learning model is constructed by machine learning using the past three-dimensional data and the reference information as learning data. An evaluation system that evaluates 3D data based on the surface condition of the evaluation object.
An acquisition means for acquiring the three-dimensional data having a plurality of position data indicating the coordinates of the three-dimensional space with respect to the evaluation object and color data indicating the color in the position data of the evaluation object.
A reference database in which a learning model having a relationship between the past 3D data acquired in advance and the reference information associated with the past 3D data is stored.
An evaluation means that refers to the reference database and generates an evaluation result for the three-dimensional data,
An evaluation system characterized by being equipped with.

Images