Machine Vision Based Traffic Sign Detection Methods: Review, Analyses and Perspectives

Abstract

Traffic signs recognition (TSR) is an important part of some advanced driver-assistance systems (ADASs) and auto driving systems (ADSs). As the first key step of TSR, traffic sign detection (TSD) is a challenging problem because of different types, small sizes, complex driving scenes, and occlusions. In recent years, there have been a large number of TSD algorithms based on machine vision and pattern recognition. In this paper, a comprehensive review of the literature on TSD is presented. We divide the reviewed detection methods into five main categories: color-based methods, shape-based methods, color-and shape-based methods, machine-learning-based methods, and LIDAR-based methods. The methods in each category are also classified into different subcategories for understanding and summarizing the mechanisms of different methods. For some reviewed methods that lack comparisons on public datasets, we reimplemented part of these methods for comparison. The experimental comparisons and analyses are presented on the reported performance and the performance of our reimplemented methods. Furthermore, future directions and recommendations of the TSD research are given to promote the development of the TSD.

A Review of Sensor Technologies for Perception in Automated Driving

“…The spatial and temporal scopes of applicability are also defined in the sign, either explicitly or implicitly. Acquiring information from road traffic signs involves two major tasks: Traffic Sign Detection (TSD) which consists on finding the location, orientation and size of traffic signs in natural scene images, and Traffic Sign Recognition (TDR) or classifying the detected traffic signs into types and categories in order to extract the information that they are providing to drivers [47].…”

Focus First: Coarse-to-Fine Traffic Sign Detection With Stepwise Learning

“…Traffic sign detection algorithms have been studied since the 1990s [30]. At first, traffic signs were detected mainly through the methods based on color and shape.…”

“…With the significant success on other computer vision tasks, machine learning strategies are also widely used in traffic sign detection problem. Before 2012, AdaBoost [48] or SVM [8] based detection methods received impressive success [30]. For example, Zaklouta et al [53] trained SVM to detect traffic targets by utilizing extracted HOG features [9] and achieved a good detection effect.…”

“…A complete intelligent traffic decision-making system needs to perceive the information of traffic signs on the road to make correct decisions. Therefore, effective detection of the road traffic signs is of great importance [30]. However, the detection technology of the traffic signs needs to overcome a series of problems such as geometric distortion caused by the shooting angle, occlusion, deformation of traffic signs and motion blur brought by the high-speed move-ment of the vehicles.…”

A Combined Object Detection Method With Application to Pedestrian Detection

“…The detection of object is very vital for automatic driving. Over the past decade, there have been significant researches on this area [1], including the detection of pedestrian [2], traffic sign [3] [4], vehicle [5] [6], traffic light [7] and etc. The objective of detection system is to provide the driver or automatic driving system with timely and accurate information about the environment to save lives by reducing the number of traffic accidents [8] [9].…”