Coating evaluation device and coating evaluation method

1 . A coating evaluation device comprising:

a shape acquisition unit including a 3D scanner, the shape acquisition unit being configured to acquire shape information representing a curved shape of a coating surface;

a surface roughness acquisition unit including a laser microscope, the surface roughness acquisition unit being configured to acquire surface roughness information representing a surface roughness of the coating surface; and

a controller including a processor, the controller being configured to

estimate an evaluation value that corresponds to a combination of the shape information and the surface roughness information by using an evaluation model that outputs a brilliance evaluation value pertaining to the coating surface in response to an input including the shape information and the surface roughness information.

2 . The coating evaluation device according to claim 1 , wherein

the evaluation model is a trained model generated through machine learning that is based on teaching data in which shape information pertaining to an evaluated coating surface, the surface roughness information pertaining to the evaluated coating surface, and the brilliance evaluation value pertaining to the evaluated coating surface.

3 . The coating evaluation device according to claim 1 , wherein

the brilliance evaluation value pertaining to the coating surface is an index determined according to at least one of smoothness of the coating surface, a proportion of light reflected by the coating surface via diffuse reflection, and resolution of an image appearing on the coating surface.

4 . The coating evaluation device according to claim 1 , wherein

the shape acquisition unit is configured to acquire design data pertaining to the coating surface as the shape information.

5 . The coating evaluation device according to any of claim 4 , wherein

the design data represents an extent of curving of the coating surface or an extent of sloping of the coating surface.

6 . The coating evaluation device according to claim 1 , wherein

the shape acquisition unit is configured to acquire measurement data obtained by measuring the coating surface as the shape information.

7 . The coating evaluation device according to claim 1 , further comprising

an image acquisition unit including a digital camera, the image acquisition unit being configured to acquire a captured image of the coating surface, and

the controller being configured to associate the captured image and the shape information with the surface roughness information and records a position on the coating surface for which the surface roughness information was acquired.

8 . The coating evaluation device according to any of claim 7 , wherein

the image acquisition unit is configured to is configured to set at least one of a focal length, an angle of view of a lens, an angle formed by a direction perpendicular to the camera and a horizontal direction.

9 . The coating evaluation device according to claim 1 , wherein

the surface roughness acquisition unit is configured to acquire measurement data obtained by measuring the coating surface as the surface roughness information.

10 . A coating evaluation method comprising:

using a controller including a processor to execute;

acquiring shape information representing a curved shape of a coating surface;

acquiring surface roughness information representing a surface roughness of the coating surface; and

estimating an evaluation value that corresponds to a combination of the shape information and the surface roughness information by using an evaluation model that outputs a brilliance evaluation value pertaining to the coating surface in response to an input including the shape information and the surface roughness information.

11 . A non-transitory computer-readable storage medium having a coating evaluation program stored thereon, the program being executable by a computer that includes a processor, the processor being configured to control

a shape acquisition unit that includes a 3D scanner and is configured to acquire shape information representing a curved shape of a coating surface, and

a surface roughness acquisition unit that includes a laser microscope and is configured to acquire surface roughness information representing a surface roughness of the coating surface,

to execute

a step for acquiring the shape information by using the shape acquisition unit,

a step for acquiring the surface roughness information by using the surface roughness acquisition unit, and

a step for estimating an evaluation value that corresponds to a combination of the shape information and the surface roughness information by using an evaluation model that outputs a brilliance evaluation value pertaining to the coating surface in response to an input including the shape information and the surface roughness information.

12 . A non-transitory computer-readable storage medium having a computer evaluation model stored thereon, the computer evaluation model being configured from a neural network that includes an input layer and an output layer, the evaluation model being trained by associating

input data that is inputted to the input layer and includes shape information representing a curved shape of a coating surface and surface roughness information representing a surface roughness of the coating surface, and

output data that is outputted from the output layer and includes a brilliance evaluation value pertaining to the coating surface.

13 . An evaluation model generation method comprising:

using a controller including a processor to execute;

acquiring teaching data in which shape information representing a curved shape of a coating surface, surface roughness information representing a surface roughness of the coating surface, and a brilliance evaluation value pertaining to the coating surface; and

performing machine learning that is based on the teaching data to generate an evaluation model that outputs the brilliance evaluation value pertaining to the coating surface in response to an input including the shape information and the surface roughness information.