Obstacle Detection Challenges of Camera Sensor Designed for ADAS

Abstract

Camera is a very crucial sensor for ADAS which is commonly used in all the vehicles to assist driver by providing information about all the obstacles around the vehicle during a drive. The Camera sensor is a vision based sensor and it is highly preferred because of its advantages like excellent in classification, good resolution, very economical and small in size. Still this sensor is having some disadvantages like huge computational load, failed to detect obstacles due to poor light and weather conditions and even less capable to estimate the distance from the obstacle. In recent years so many research works were performed to overcome these challenges, for making the camera a robust sensor. In this paper, the challenges in detection faced by camera sensor were focused and their respective solution is discussed, which are required to improve the effectiveness in vision based detection.