Abstract: Automatic detection, detection of traffic, Lane and traffic are important and may be used to help drivers reduce errors and ultimately in self driving vehicles. In this paper, the CNN model is used to create an Autonomous Traffic sign, lane and traffic recognition system. The proposed framework works continuously to distinguish and visualise traffic signals, identify lanes by hauge transform and differentiation between vehicles using canny. The commitment of this paper is in addition to the newly acquired knowledge of 43 different road signs collected on the sides of unusual locations, lanes, cars and various trucks in India. Images are taken at various points and include various parameters and conditions. A total of 40000+ images were collected to include a set of data sets we named Indian Traffic and road signs. In cars and traffic we use a CNN model called YOLO. CNN engineering has been used with flexible parameters to achieve the best accuracy results. Test results show that the proposed CNN engineering has 98% accuracy, in this way higher than that achieved in previous comparative studies.

Keywords: Lane cars sign Detection, Convolutional Neural Network, Deep Learning, Open CV, Canny,

| DOI: 10.17148/IARJSET.2022.9518