A Review of Automatic Road Sign Recognition Systems

Abstract

Traffic-signs recognition(TSR) systems are an in-herent element of Intelligent Transport Systems (ITS), which is why it's considered a necessary element of Advanced Driver Assistance Systems (ADAS). One of our biggest challenges for driver assistance systems requires realizing the surrounding environment while driving the car. This paper details standard Traffic-signs recognition(TSR) system methods used in recent research. Features and classifiers used in modern TSR system methods are introduced and discussed. Most of these methods achieve great precision (accuracy of 98%) when tested in the German Traffic Sign Recognition Benchmark (GTSRB), the most extensively employed and publicly accessible TSR dataset. The paper compares the latest techniques by evaluating their processing time and precision. Furthermore, a discussion of possible future work is provided.