This project aims to provide the facility of an Automated Driving Test in order to assess the skills and capabilities of drivers applying for a license. Normally, an applicant lacking proper driving skills and road sense is also granted a driving license due to human negligence and unethical practices.
Therefore, the proportion of road accidents is increasing due to these unqualified drivers. For the minimization of this problem automation of the driving test is the key and to improve the test the emotions of the driver will also be monitored.
Machine Learning and Computer Vision algorithms will be used so that an efficient, transparent, and reliable system can be built that permits only skilled drivers having knowledge of traffic rules to have a driver’s license, but the computer vision and machine learning algorithms being used for Line Crossing detection, Car detection, and Emotion detection and classification are highly computationally expensive so it won’t be possible to develop a system simply using already developed algorithms.
Therefore, we are going to establish a system with minimum acceptable accuracy loss as a tradeoff for enhanced performance and less computational burden.