TIRE WEAR DEGREE ESTIMATION DEVICE, TIRE WEAR DEGREE LEARNING DEVICE, TIRE WEAR DEGREE ESTIMATING METHOD, AND LEARNED MODEL PRODUCING METHOD AND PROGRAM
To provide a tire wear degree estimating device capable of accurately estimating the wear degree of various kinds of tires even if data amount used for learning is small, a tire wear degree learning device, a tire wear degree estimating method, and a learned model producing method and a program.SOLU...
Gespeichert in:
Hauptverfasser: | , |
---|---|
Format: | Patent |
Sprache: | eng ; jpn |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | To provide a tire wear degree estimating device capable of accurately estimating the wear degree of various kinds of tires even if data amount used for learning is small, a tire wear degree learning device, a tire wear degree estimating method, and a learned model producing method and a program.SOLUTION: A tread surface image obtaining portion 64 obtains a tread surface image showing a tread surface of a tire. A tread surface interest region specifying portion 66 specifies a tread surface interest region showing a tread surface slip sign in the tread surface image. A target input image generating portion 68 generates a target input image that is an image of the tread surface interest region of predetermined pixels. An estimating portion 70 estimates a wear degree of a tire, on the basis of output when the target input image is inputted in a mechanical learning model 50 that has been learned.SELECTED DRAWING: Figure 7
【課題】学習に用いられるデータ量が少なくても様々な種類のタイヤの摩耗度を的確に推定できるタイヤ摩耗度推定装置、タイヤ摩耗度学習装置、タイヤ摩耗度推定方法、学習済モデルの生成方法及びプログラムを提供する。【解決手段】トレッド面画像取得部64は、タイヤのトレッド面が写るトレッド面画像を取得する。トレッド面着目領域特定部66は、トレッド面画像においてトレッド面スリップサインが写るトレッド面着目領域を特定する。ターゲット入力画像生成部68は、所定画素数のトレッド面着目領域の画像であるターゲット入力画像を生成する。推定部70は、ターゲット入力画像を学習済の機械学習モデル50に入力した際の出力に基づいて、タイヤの摩耗度を推定する。【選択図】図7 |
---|