Image Classification for Advanced Driving Assistant System in Electric Wheelchair

Abstract

Demands for welfare equipment such as electric wheelchairs are increasing in Japan, because of traffic accidents by the elderly. In the field of automobiles, the safety driving supports have been researched in many divergences. However, in the field of electric wheelchairs, not advanced too much. In this study, we propose a method of machine learning by Self-Supervised Learning (SSL) and a method of image classification according to the results, using actual sidewalk images as a machine learning dataset, to Advanced Driving Assistant System (ADAS) of electric wheelchairs.