Real-time traffic sign recognition system

Abstract

Traffic sign recognition (TSR) is one of the most important background research topics for enabling autonomous vehicle driving systems. Autonomous driving systems require special handling of input data: there is no time for complex transformations or sophisticated image processing techniques, they need a solid and real-time analysis of a situation. This challenge get more difficult to meetin a city like environment where multiple traffic signs, ads, parking vehicles, pedestrians, and other moving or background objects make the recognition much more difficult. While numerous solutions have been published, solutions are tested on autoways, country-side, or at a very low speed. In this paper, we give a short overview on main problems and known strategies to solve these problems, and we give a general solution to tackle real-time issues in urban traffic sign recognition.