Advanced Driving Assistance System using Computer Vision and Deep Learning Algorithms

Advanced Driving Assistant System (ADAS) is one of the key safety features in every modern car as most car accidents happen due to driver error. By using ADAS, this kind of accident can be either avoided or the damage can be minimized. This research study intends to develop an ADAS system using Computer Vision (CV) and Deep Learning (DL) algorithm, which can be implemented in lane detection, lane keeping assist, autonomous braking assist, high beam assist, and blind spot monitoring of the vehicles. For lane keeping assist and lane detection, CV is used whereas for the other safety features, object detection model has been built by using DL. For the object detection model, the precision and mean average precision of the overall model has been analyzed to enhance the performance of the proposed system.