A Novel Traffic Sign Recognition System Combining Viola–Jones Framework and Deep Learning

Abstract

An automated traffic sign detection system is an important aspect of advanced driver assistance systems (ADAS) as it significantly enhances the driver’s situational awareness. In this paper, a novel approach combining Viola–Jones framework and deep learning, to design of a camera-based traffic sign recognition system capable of detecting and identifying the traffic signs, is presented. The proposed system is designed and tested for its effectiveness under Indian road conditions. It is observed that the designed system is well capable of correctly detecting the correct traffic signs even in the presence of non-standardized traffic signs.