JP7462480B2 - Tire performance prediction model learning method, tire performance prediction method, system, and program - Google Patents

Description

The present disclosure relates to a tire performance prediction model learning method, a tire performance prediction method, a system, and a program.

The use of machine learning is spreading to various industrial fields, and there are also increasing examples of other companies in the same industry applying it to tires. For example, Patent Document 1 describes how images of tire treads are input into a neural network to estimate the state or amount of wear.

The inventors of the present disclosure believe that tire performance is reflected in the contact patch because tire designers often discuss the tire based on the tire contact patch. Therefore, if tire performance values could be predicted from a contact patch image that represents the shape of the contact patch of a tire that contacts the road surface under a specified load, this would be useful as it would lead to a reduction in the number of prototypes and tests. It would also provide a clue as to which elements of the contact patch are related to each tire performance. However, no specific method has been proposed for predicting tire performance from a tire contact patch image.

JP 2019-35626 A

The present disclosure provides a learning method, a tire performance prediction method, a system, and a program for a tire performance prediction model for predicting tire performance values based on a tire contact patch image.

The tire performance prediction model learning method disclosed herein is a method executed by one or more processors, and includes a feature extraction step of inputting an input image based on a contact patch image representing the tire contact patch shape to a feature extraction unit to extract image features, and a learning step of machine learning a prediction model to output tire performance values using the extracted image features as input.

FIG. 1 is a diagram showing a usage manner of a tire performance prediction model learning system and a tire performance prediction system according to a first embodiment. 1 is a block diagram showing a tire performance prediction model learning system and a tire performance prediction system according to a first embodiment; 4 is a flowchart executed by the tire performance prediction model learning system of the first embodiment. 3 is a flowchart executed by the tire performance prediction system of the first embodiment. FIG. 13 is an explanatory diagram relating to a process of generating an input image by trimming a ground plane image. FIG. 13 is a block diagram showing a tire performance prediction model learning system and a tire performance prediction system according to a second embodiment. 10 is a flowchart executed by a tire performance prediction model learning system according to a second embodiment. 10 is a flowchart executed by a tire performance prediction system according to a second embodiment.

First Embodiment
Hereinafter, a first embodiment of the present disclosure will be described with reference to the drawings.

Figure 1 is a diagram showing how a tire performance prediction model learning system 1 and a tire performance prediction system 2 are used. As shown in Figure 1, a contact patch image 5 is obtained from a test device 3 or a tire contact simulation system 4. The tire performance prediction model learning system 1 trains a prediction model by machine learning using training data that associates the contact patch image 5 as an input to the prediction model with a correct tire performance value as an output from the prediction model. The input image used for the training data is based on the contact patch image. The tire performance prediction system 2 uses the prediction model constructed by the tire performance prediction model learning system 1 to calculate (predict) and output a tire performance value based on the contact patch image 5 to be predicted.

The tire performance values to be predicted are actual measured values. The tire performance values used in this embodiment are CP (cornering power), SAP (self-aligning power), CFmax (maximum cornering force), and SAT (self-aligning torque). Of course, the tire performance values can be applied to any other tire performance.

The ground contact surface image 5 represents the shape of the tire ground contact surface. The ground contact surface image 5 of this embodiment represents the shape of the tire ground contact surface and also represents the ground contact pressure Pz in a direction perpendicular to the road surface by color in the case of a color image, and by brightness in the case of a grayscale image. The ground contact surface image 5 of this embodiment is an image of the tire in a stationary state where the tire is not rolling, but is not limited to this, and may be an image of the tire rolling. Regardless of whether the tire is in a rolling state or stationary state, the camber angle of the tire can be set to any angle. The ground contact surface image 5 of this embodiment is an image of the tire in a stationary state, but when the tire is in a rolling state, the slip angle with respect to the traveling direction may be 0 degrees or may be an angle other than 0 degrees. The ground contact surface image 5 of this embodiment represents only the ground contact surface pressure Pz in a direction perpendicular to the road surface in a stationary state, but is not limited to this. For example, when the tire is in a rolling state, the pressure Px along the traveling direction of the tire or the pressure Py along a direction perpendicular to the traveling direction of the tire may be represented. Note that these coordinate systems are only examples and can be changed as appropriate. If accuracy can be ensured, the contact patch image 5 may represent only the tire contact patch shape. The test device 3 places the test target tire 30 on the test road surface 31 under a predetermined load, and captures the contact patch shape with the camera 32 through the transparent road surface of the test road surface 31. The contact pressure is measured by an optical method using a transparent plate or the like, or by a pressure sensor. The test device 3 generates the contact patch image 5 by the above test. The contact patch image 5 may also be obtained from the simulation results obtained by the tire contact simulation system 4 shown in FIG. 1.

Depending on the implementation method, the tire performance prediction model learning system 1 and the tire performance prediction system 2 may not be built on the same computer system, but may be operated independently. In other words, only the tire performance prediction model learning system 1 may be implemented, or only the tire performance prediction system 2 may be implemented.

[Tire performance prediction model learning system 1]
Fig. 2 is a block diagram showing a tire performance prediction model learning system 1 and a tire performance prediction system 2. As shown in Fig. 2, the tire performance prediction model learning system 1 includes a conversion unit 10, a feature extraction unit 11, and a learning unit 12. The conversion unit 10 is optional.

These units 10 to 12 are realized by software and hardware working together when processor 1a executes the processing routine shown in FIG. 3, which is pre-stored in a computer equipped with processor 1a, memory 1b, various interfaces, etc. In this embodiment, the processor 1a in one device realizes each unit, but this is not limited to this. For example, the units may be distributed using a network, with multiple processors executing the processing of each unit. In other words, one or multiple processors execute the processing.

When an input image 6 based on the ground surface image 5 is input, the feature extraction unit 11 extracts the feature of the image. In a configuration in which the conversion unit 10 is not provided, the ground surface image 5 is input to the feature extraction unit 11 as the input image 6. In a configuration in which the conversion unit 10 is provided, the feature extraction unit 11 inputs the input image 6 output by the conversion unit 10. The feature extraction unit 11 may use any algorithm as long as it can extract features from an image. Examples include a neural network (e.g., Alexnet, GoogleNet, ResNet101), discrete cosine transform processing, AutoEncoder, and a configuration that combines discrete cosine transform processing and AutoEncoder.

The learning unit 12 uses the teacher data set D1 (or D2) to construct a prediction model 13 by machine learning so as to output a tire performance value using the feature amount of the image extracted by the feature extraction unit 11 as an input. The teacher data set D1 is data in which the ground contact surface image 5 (ground contact surface images 1 to N) or the input image 6 (input images 1 to N) is associated with the tire performance (X ₁ , X ₂ , ..., X _N ) value. The teacher data set D2 is data in which the input image 6 (input images 1 to N) is associated with the tire performance (X ₁ , X ₂ , ..., X _N ) value. N indicates the number of teacher data. The prediction model 13 can use various models such as Gaussian process regression, linear regression, classification tree, random forest, support vector machine, and ensemble tree as long as it is a supervised machine learning model.

The conversion unit 10 converts the ground plane image 5 into an input image 6 of a size that can be input to the feature extraction unit 11. In this embodiment, the ground plane image 5, which is 875 x 656 pixels and is color or grayscale (8 bit), is converted into an input image 6, which is 227 x 227 pixels and is grayscale (8 bit), but this is an example and is not limited to this.

When converting to a size that can be input to the feature extraction unit 11, care must be taken not to destroy the image features that appear in the original contact patch image 5. The size of the contact patch affects performance. For example, the contact patch of a large tire appears large in the image, and conversely, the contact patch of a small tire appears small in the image. If the images are trimmed so that the size of the contact patch occupies the same amount after trimming, there is a risk that the performance value of small tires will be overestimated as being similar to that of large tires, when in fact the performance value should be predicted to be small. Therefore, since the size of the contact patch shape affects tire performance values, it is necessary to maintain the scale between multiple images.

To achieve this, the conversion unit 10 has a selection unit 10a, a trimming position determination unit 10b, a trimming unit 10c, and a size change unit 10d. As a premise, all of the contact patch images 5 are taken under the same shooting conditions and have the same scale. The same shooting conditions mean that the distance from the camera 32 to the test road surface 31 is the same, and the zoom value of the camera 32 is the same, as shown in FIG. 1.

The selection unit 10a selects the contact patch image with the largest contact patch shape from the teacher data set D1 that associates the contact patch images 5 with tire performance (X1 _{to N} ) and is used for machine learning. Among the contact patch images 50, 51, and 52 shown in FIG. 5, the contact patch image 50 has the largest contact patch shape.

As shown in FIG. 5, the trimming position determination unit 10b determines a trimming position P1 in the image based on the contact patch image 50 selected by the selection unit 10a. The larger the contact patch shape in the input image 6, the higher the prediction accuracy. Therefore, in this embodiment, the trimming position P1 is the minimum rectangle that includes the contact patch shape or a range obtained by expanding the minimum rectangle by a specified number of pixels. The trimming position P1 determined by the trimming position determination unit 10b is stored in the memory 1b for use by the tire performance prediction system 2.

The trimming unit 10c uses the trimming position P1 determined by the trimming position determination unit 10b to trim all of the contact surface images 5 (images 1 to N) in the teacher dataset D1 to generate trimmed images.

The resizing unit 10d resizes each trimmed image generated by the trimming unit 10c to a size that can be input to the feature extraction unit 11, generating an input image 6. The resizing unit 10d resizes the trimmed image without changing its aspect ratio. In the example of FIG. 5, the input image 60 is generated from the ground plane image 50 via a trimmed image (not shown). The input image 61 is generated from the ground plane image 51 via a trimmed image (not shown). The input image 62 is generated from the ground plane image 52 via a trimmed image (not shown). By performing such processing, it is possible to prevent the scale of the ground plane shape depicted in each of the input images 60 to 62 from becoming inconsistent.

The tire performance with improved prediction performance as a result of trimming by the conversion unit 10 while maintaining the aspect ratio within one image and the relative size (scale) between multiple images is as follows.
Cornering: CP, SAP, CFmax (maximum cornering force), SAT
Traction system: Braking performance on dry surfaces, braking performance on wet surfaces, braking performance on icy surfaces, braking performance on snowy surfaces, friction performance on ice, friction performance on snow, hydroplaning resistance Noise system: Noise performance inside and outside the vehicle, sound radiation performance in individual tire tests Rolling resistance Abrasion resistance Heel-toe wear performance and uneven wear performance are influenced by the ratio and distribution of areas of high contact pressure within the contact patch and the shape of the contact patch, and therefore trimming or size changes that retain the above tire size information are not considered necessary.

[Tire performance prediction system 2]
2, the tire performance prediction system 2 includes a conversion unit 20, a feature extraction unit 21, and a prediction unit 22. The conversion unit 20 is optional, similar to the tire performance prediction model learning system 1.

When an input image 6 based on the contact patch image 5 is input, the feature extraction unit 21 extracts image features. In a configuration in which the conversion unit 20 is not provided, the contact patch image 5 is input to the feature extraction unit 21 as the input image 6. In a configuration in which the conversion unit 20 is provided, the input image 6 output by the conversion unit 20 is input. The feature extraction unit 21 in the tire performance prediction system 2 has the same configuration as the feature extraction unit 11 in the tire performance prediction model learning system 1.

The prediction unit 22 uses the prediction model 13 constructed by the tire performance prediction model learning system 1 to input the image features output by the feature extraction unit 21 and output tire performance values.

As shown in FIG. 2, the conversion unit 20 converts the contact surface image 5 to be predicted into an input image 6 suitable for input to the feature extraction unit 21. The conversion unit 20 has a trimming unit 20c and a size change unit 20d. The trimming unit 20c trims the contact surface image 5 using a predetermined trimming position P1 to generate a trimmed image. The trimming position P1 is determined by the trimming position determination unit 10b and stored in the memory 1b. The size change unit 20d changes the size of the trimmed image generated by the trimming unit 20c to a size that can be input to the feature extraction unit 21 to generate an input image 6. The size change unit 20d changes the size without changing the aspect ratio of the trimmed image. The trimming unit 20c has the same configuration as the trimming unit 10c in the tire performance prediction model learning system 1. The size change unit 20d has the same configuration as the size change unit 10d in the tire performance prediction model learning system 1.

[Training method for tire performance prediction model]
A tire performance prediction model learning method executed by one or more processors in the tire performance prediction model learning system 1 shown in FIG. 2 will be described with reference to FIG.

First, by executing steps ST1 to ST4, the conversion unit 10 generates an input image 6 based on the contact patch image 5. Specifically, in step ST1, the selection unit 10a selects the contact patch image 50 having the largest contact patch shape from the teacher data set D2 used in machine learning, which associates the contact patch images 5 (images 1 to N) with tire performance (X _{1 to N} ).
In the next step ST2, the trimming position determination unit 10b determines a trimming position P1 in the image based on the selected contact patch image 50.
In the next step ST3, the trimming unit 10c uses the trimming position P1 determined by the trimming position determination unit 10b to trim all the contact patch images (1 to N) of the teacher data set D1 to generate trimmed images.
In the next step ST4, the size changing unit 10d changes the size of each trimmed image to a size that can be input to the feature extraction unit 11 without changing the aspect ratio of each trimmed image, thereby generating an input image 6 (60-62).
In the next step ST5, the feature extraction unit 11 inputs an input image 6 based on the contact patch image 5 representing the tire contact patch shape to the feature extraction unit 11, and extracts the feature amount of the image.
In the next step ST6, the learning unit 12 trains the prediction model 13 through machine learning so as to output tire performance values using the extracted image features as input.

[Tire performance prediction method]
A tire performance prediction method executed by one or more processors in the tire performance prediction system 2 shown in FIG. 2 will be described with reference to FIG.

First, by executing steps ST101 to ST102, the conversion unit 20 generates an input image 6 based on the ground plane image 5. Specifically, in step ST101, the trimming unit 20c uses a predetermined trimming position P1 to trim the ground plane image 5 to be predicted, thereby generating a trimmed image.
In the next step ST102, the size change unit 20d changes the size of the trimmed image to a size that can be input to the feature extraction unit 21 without changing the aspect ratio of the trimmed image, thereby generating an input image 6.
In the next step ST103, the feature extraction unit 21 inputs the input image 6 based on the contact patch image 5 representing the tire contact patch shape to the feature extraction unit 21, and extracts the feature amount of the image.
In the next step ST104, the prediction unit 22 outputs a tire performance value corresponding to the extracted image feature quantity using a prediction model 13 that has been machine-trained to output a tire performance value using the extracted image feature quantity as input.

Second Embodiment
Hereinafter, a second embodiment of the present disclosure will be described with reference to the drawings.

As shown in FIG. 6, the tire performance prediction model learning system 1 and tire performance prediction system 2 of the second embodiment are configured to perform the trimming process using data D3 that previously associates tire size with trimming position.

6 and 7, the conversion unit 10 of the tire performance prediction model learning system 1 of the second embodiment has a trimming position specification unit 10e. The memory 1b stores data D3 that associates tire sizes with trimming positions.
The operation of the tire performance prediction model learning system 1 is as follows.
First, by executing steps ST201 to ST203, the conversion unit 10 generates the input image 6 based on the contact patch image 5. Specifically, in step ST201, the trimming position specifying unit 10e specifies a trimming position corresponding to a specified tire size based on data D3 that previously associates tire sizes with trimming positions.
In the next step ST202, all the contact patch images (1 to N) of the teacher data set D1 are trimmed using the trimming position P1 specified by the trimming position specifying unit 10e to generate respective trimmed images.
In the next step ST203, the size changing unit 10d changes the size of each trimmed image to a size that can be input to the feature extraction unit 11 without changing the aspect ratio of each trimmed image, thereby generating an input image 6 (60-62).
In the next step ST204, the feature extraction unit 11 inputs the input image 6 based on the contact patch image 5 representing the tire contact patch shape to the feature extraction unit 11, and extracts the feature amount of the image.
In the next step ST205, the learning unit 12 trains the prediction model 13 by machine learning so as to output tire performance values using the extracted image feature amounts as input.

6 and 8, the conversion unit 20 of the tire performance prediction system 2 of the second embodiment has a trimming position specification unit 20e. The memory 1b stores data D3 associating tire sizes with trimming positions.
The operation of the tire performance prediction system 2 is as follows.
First, by executing steps ST301 to ST303, the conversion unit 20 generates the input image 6 based on the contact patch image 5. Specifically, in step ST301, the trimming position specifying unit 20e specifies the trimming position corresponding to the specified tire size based on data D3 that previously associates tire sizes with trimming positions.
In the next step ST302, the trimming unit 20c uses the trimming position P1 specified by the trimming position specifying unit 20e to trim the contact patch image 5 to be predicted, thereby generating a trimmed image.
In the next step ST303, the size change section 20d changes the size of the trimmed image to a size that can be input to the feature extraction section 21 without changing the aspect ratio of the trimmed image, thereby generating an input image 6.
In the next step ST304, the feature extraction unit 21 inputs the input image 6 based on the contact patch image 5 representing the tire contact patch shape to the feature extraction unit 21, and extracts the feature amount of the image.
In the next step ST305, the prediction unit 22 outputs a tire performance value corresponding to the extracted image feature quantity using a prediction model 13 that has been machine-trained to output a tire performance value using the extracted image feature quantity as input.

As described above, the learning method of the tire performance prediction model of the first or second embodiment is a method executed by one or more processors, and includes a feature extraction step of inputting an input image 6 based on a contact patch image 5 representing the tire contact patch shape to a feature extraction unit 11 to extract image features, and a learning step of machine learning the prediction model 13 to output tire performance values using the extracted image features as input.

The tire performance prediction model learning system 1 of the first or second embodiment includes a feature extraction unit 11 that inputs an input image 6 based on a contact patch image 5 representing the tire contact patch shape and extracts image features, and a learning unit 12 that trains a prediction model 13 by machine learning so as to output tire performance values using the extracted image features as input.

This makes it possible to provide a prediction model 13 that predicts tire performance values based on features extracted from an input image 6 based on the contact patch image 5, making it possible to know tire performance values based on the contact patch image 5.

The tire performance prediction method of the first or second embodiment is a method executed by one or more processors, and includes a feature extraction step of inputting an input image 6 based on a contact patch image 5 representing the tire contact patch shape to a feature extraction unit 21 to extract image features, and a prediction step of outputting tire performance values corresponding to the extracted image features using a prediction model 13 that has been machine-learned to input the extracted image features and output tire performance values.

The tire performance prediction system 2 of the first or second embodiment includes a feature extraction unit 21 that receives an input image 6 based on a contact patch image 5 representing the tire contact patch shape and extracts image features, and a prediction unit 22 that uses a prediction model 13 that has been machine-learned to input the extracted image features and output tire performance values, and outputs tire performance values corresponding to the extracted image features.

As a result, the prediction model 13 predicts the tire performance value based on the feature values extracted from the input image 6 based on the contact patch image 5, making it possible to know the tire performance value based on the contact patch image 5.

Although not particularly limited, as in the first or second embodiment, it is preferable that the contact patch image 5 represents the tire contact patch shape and contact pressure.

Although not particularly limited, as in the learning method of the tire performance prediction model of the first embodiment, the method may include an image generation step of generating an input image 6 based on a contact patch image 5, and the image generation step may include a step of selecting a contact patch image 50 having the largest contact patch shape from the teacher data set D1 that associates the contact patch image 5 with the tire performance value and is used in the learning step, a step of determining a trimming position P1 in the image based on the selected contact patch image 50, a step of trimming all the contact patch images 5 of the teacher data set D1 using the determined trimming position P1 to generate trimmed images, and a step of changing the size of each trimmed image to a size that can be input to the feature extraction unit 11 without changing the aspect ratio of each trimmed image to generate an input image 6.

As a result, the contact patch image 5 with the largest contact patch shape is selected to determine the trimming position P1, making it possible to make the contact patch occupy as large a portion of the trimmed image as possible, thereby improving the prediction accuracy of the prediction model 13. At the same time, the determined trimming position is used in common for all contact patch images 5 for trimming, making it possible to correctly predict tire performance values by taking into account differences in the size of the contact patch shape due to tire size.

Although not particularly limited, like the tire performance prediction method of the first embodiment, the method may include an image generation step of generating an input image 6 based on the contact patch image 5, and the image generation step may include a step of generating a trimmed image by trimming the contact patch image 5 using a predetermined trimming position, and a step of changing the size of the trimmed image to a size that can be input to the feature extraction unit 21 without changing the aspect ratio of the trimmed image, thereby generating the input image 6.

As a result, trimming is performed using a common, predefined trimming position, making it possible to accurately predict tire performance values while taking into account differences in the size of the contact patch shape due to tire size.

Although not particularly limited, like the tire performance prediction model learning method or tire performance prediction method of the second embodiment, it may include an image generation step of generating an input image 6 based on a contact patch image 5, and the image generation step may include a step of identifying a trimming position corresponding to a specified tire size based on data D3 that previously associates tire sizes with trimming positions, a step of generating a trimmed image by trimming the contact patch image 5 using the identified trimming position, and a step of changing the size of the trimmed image to a size that can be input to the feature extraction units 11 and 21 without changing the aspect ratio of the trimmed image to generate the input image 6.

This allows trimming to be performed using common or separate appropriate trimming positions according to tire size, making it possible to accurately predict tire performance values while taking into account differences in the size of the contact patch shape due to tire size.

The program according to this embodiment is a program for causing a computer to execute the above method.
By executing these programs, it is possible to obtain the effects of the above-mentioned methods.

Although the embodiments of the present disclosure have been described above with reference to the drawings, the specific configuration should not be considered to be limited to these embodiments. The scope of the present disclosure is indicated not only by the description of the above embodiments but also by the claims, and further includes all modifications within the meaning and scope equivalent to the claims.

The structures employed in each of the above embodiments can be employed in any other embodiment. The specific configurations of each part are not limited to the above-mentioned embodiments, and various modifications are possible without departing from the spirit of this disclosure.

For example, the order of execution of each process, such as operations, procedures, steps, and stages, in the devices, systems, programs, and methods shown in the claims, specifications, and drawings, can be realized in any order, so long as the output of a previous process is not used in a subsequent process. Even if the flow in the claims, specifications, and drawings is explained using "first," "next," etc. for convenience, this does not mean that it is necessary to execute the processes in this order.

Each unit shown in Figures 2 and 7 is realized by executing a specific program on one or more processors, but each unit may also be configured with a dedicated memory or dedicated circuit. In the system 1 of the above embodiment, each unit is implemented in the processor 1a of one computer, but each unit may be distributed and implemented on multiple computers or in the cloud. In other words, the above method may be executed on one or more processors.

System 1 includes a processor 1a. For example, processor 1a can be a central processing unit (CPU), a microprocessor, or other processing unit capable of executing computer-executable instructions. System 1 also includes memory 1b for storing data for system 1. In one example, memory 1b includes computer storage media, including RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, DVD or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other media that can be used to store desired data and that can be accessed by system 1.

1... Tire performance prediction model learning system, 10... Conversion unit, 11... Feature extraction unit, 12... Learning unit, 2... Tire performance prediction system, 20... Conversion unit, 21... Feature extraction unit, 22... Prediction unit

Claims (8)

A method executed by one or more processors, comprising:
a feature extraction step of inputting an input image based on a contact patch image representing a tire contact patch shape to a feature extraction unit and extracting a feature of the image;
a learning step of machine learning a prediction model so as to output a tire performance value using the extracted image feature quantity as an input;
A method for training a tire performance prediction model, including: A method executed by one or more processors, comprising:
a feature extraction step of inputting an input image based on a contact patch image representing a tire contact patch shape to a feature extraction unit and extracting a feature of the image;
a prediction step of outputting a tire performance value corresponding to the extracted image feature value by using a prediction model that has been machine-learned to output a tire performance value using the extracted image feature value as an input;
A tire performance prediction method comprising: The method according to claim 1 or 2, wherein the contact patch image represents the tire contact patch shape and contact pressure. an image generating step of generating the input image based on the ground plane image;
The image generating step includes:
selecting a contact patch image having the largest contact patch shape from a teacher data set used in the learning step, the teacher data set associating the contact patch image with the tire performance value;
determining a cropping position in the image based on the selected ground plane image;
cropping the ground plane images of the teacher data set using the determined cropping positions to generate respective cropped images;
generating the input image by changing the size of each of the trimmed images to a size that can be input to the feature extraction unit without changing the aspect ratio of each of the trimmed images;
The method of claim 1 , comprising: an image generating step of generating the input image based on the ground plane image;
The image generating step includes:
cropping the ground plane image using a predetermined cropping location to generate a trimmed image;
generating the input image by changing the size of the trimmed image to a size that can be input to the feature extraction unit without changing the aspect ratio of the trimmed image;
The method of claim 2 , comprising: an image generating step of generating the input image based on the ground plane image;
The image generating step includes:
Identifying a trimming position corresponding to a specified tire size based on data in which tire sizes and trimming positions are previously associated with each other;
cropping the ground plane image using the identified cropping locations to generate a trimmed image;
generating the input image by changing the size of the trimmed image to a size that can be input to the feature extraction unit without changing the aspect ratio of the trimmed image;
The method of claim 1 or 2, comprising: A system comprising one or more processors for executing the method according to any one of claims 1 to 6. A program causing one or more processors to execute the method according to any one of claims 1 to 6.

Images