Recognition and Classification of Traffic Signs using Machine Learning Techniques

Abstract

The computerized recognition and classification of traffic signs is a challenging problem, with several important request areas, including advanced drivers assistance systems, autonomous vehicles and street surveying. While much research is present on both automated diagnosis and popularity of symbol-based traffic indicators there is much less research concentrated specifically on the reputation of wording on traffic information indications. This may be partial because of the difficulty of the duty brought on by problems, such as brightness and shadows, blurring, occlusion, and signal deterioration. Our method of this issue by detecting many text-based traffic indication prospects using basic condition and color information. The proposed system includes two main periods: Recognition and Classification. The Acceptance stage exploits the understanding of the composition of the Traffic indication, i.e., the condition and size of the sign in the frame, to look for the locations in the landscape that it will seek out traffic text indications.