A Study on Traffic Sign Detection and Recognition Systems Using Machine Learning

Abstract

In recent times, Advanced Driver Assistance Systems (ADAS) and Autonomous Driving Systems (ADS) have become the major research areas in the automobile sector. Further, Traffic Sign Recognition (TSR) plays a significant role in ADAS and ADS. Therefore, TSR can be classified in two stages: Detection and Recognition through classification. In the traffic system, detecting traffic signs is tricky because of the signs’ shapes, sizes, and complexity of the scenes in which they appear. Moreover, the vision parameters also play a vital role. Certainly, we have seen significant developments in machine learning (ML) and deep learning (DL) methodologies in recent years. Eventually, it has become the mainstream method that helps to detect and classify signs and has achieved phenomenal results. In this paper, we have reviewed and evaluated several ML-based methods for TSR with their performance.