Tire state estimation method
AssigneeSumitomo Rubber Industries, Ltd.
InventorMariko Iwata
Life of this patent
Estimating the term…
Abstract
Provided is a method, etc., for estimating uneven tire wear. This tire state estimation method involves: obtaining an image which includes both ends of the tread of a target tire, and captures said tread from the front in a manner such that said tread is continuous in a prescribed direction; inputting the obtained image into a trained first machine learning model; and deriving an output from the trained first machine learning model. Therein, the output of the trained first machine learning model corresponds to the estimation results for the uneven wear of the target tire.
Claims15
A tire state estimation method comprising:
- acquiring an image of a tread of a target tire from a front side, the image being obtained such that the image includes both ends of the tread and the tread is continuous in a predetermined direction;
- inputting the acquired image to a trained first machine learning model;
- deriving an output from the trained first machine learning model;
- determining whether or not the target tire is required to be replaced based on the output derived from the trained first machine learning model;
- inputting the acquired image to a trained second machine learning model that is different from the first machine learning model;
- and
- deriving an output from the trained second machine learning model,
- wherein the output from the trained first machine learning model corresponds to an estimation result of uneven wear of the target tire,
- wherein the output from the trained second machine learning model corresponds to an estimation result of a depth of a groove of the tread of the target tire, and
- wherein the determining whether or not the target tire is required to be replaced is determining whether or not the target tire is required to be replaced based on the output derived from the trained first machine learning model and the output derived from the trained second machine learning model.
A tire state estimation device comprising:
- an image acquirer that acquires an image of a tread of a target tire from a front side, the image being obtained such that the image includes both ends of the tread and the tread is continuous in a predetermined direction;
- a memory that stores a trained first machine learning model;
- and
- a deriver that inputs the acquired image to the trained first machine learning model, and derives an output from the trained first machine learning model,
- wherein the output from the trained first machine learning model corresponds to an estimation result of uneven wear of the target tire,
- wherein the memory further stores a trained second machine learning model that is different from the trained first machine learning model,
- the deriver further inputs the acquired image to the trained second machine learning model, and further derives an output from the trained second machine learning model, and
- the output from the trained second machine learning model corresponds to an estimation result of a depth of a groove of the tread of the target tire.
A non-transitory computer-readable medium storing a tire state estimation program that causes a computer to execute the steps comprising:
- acquiring an image of a tread of a target tire from a front side, the image being obtained such that the image includes both ends of the tread and the tread is continuous in a predetermined direction;
- inputting the acquired image to a trained first machine learning model;
- and
- deriving an output from the trained first machine learning model;
- inputting the acquired image to a trained second machine learning model that is different from the trained first machine learning model;
- deriving an output from the trained second machine learning model;
- and
- determining whether or not the target tire is required to be replaced based on the output derived from the trained first machine learning model and the output derived from the trained second machine learning model,
- wherein the output from the trained first machine learning model corresponds to an estimation result of uneven wear of the target tire, and
- wherein the output from the trained second machine learning model corresponds to an estimation result of a depth of a groove of the tread of the target tire.
Dependencies are inferred from the claim text.