人間の視覚特性を模倣した自律移動ロボットのための画像処理法に関する提案

Jazyk: japonština
Rok vydání: 2015
Předmět:
Zdroj: 法政大学大学院紀要. 理工学・工学研究科編. 56:1-8
ISSN: 2187-9923
Popis: Human color sense is different from camera captured color image due to the characteristic human visual sensation of optical illusion and color constancy. Regardless of optical illusion and color constancy, human car-driver can handle an automobile without danger of overturning. In this paper, we describe a new white lane detection algorithm based on similar method of human color visual sensation for autonomous mobile robot. In order to safely drive on track lane guiding course, we emulate human color visual sensation to reduce the effects of lighting and shadows on the course. Validity of proposed new image compensation methods is confirmed by actual while line detection.
Databáze: OpenAIRE