Color Segmentation Of Traffic Sign In Complex Background Environment

The rapid development of economy, the continuous improvement of living standards, let the car into the household. Cars to bring convenience people trip at the same time, also caused the traffic congestion, accidents frequently. So, how to improve the security of driving a car is the biggest problems of the traffic management at present. The research of intelligent transportation technologies is also expanded in this premise. And with traffic signs for identification of the object traffic signs recognition system is an important branch of the Intelligent Transportation System. The role of traffic signs basically is to give drivers prompting at present road traffic information, warning of the potential danger of road, so as to improve the safety of driving, reduce accidents.Traffic signs in real complex scenes as the object of study in this paper, by specific algorithm for the first step of the identification of the traffic signs, that is the segmentation of the traffic signs. Because of the segmentation results has a direct impact on the sign of recognition, so can the realization of high speed and efficiency of the sign of division, has become the key to the success of the traffic sign recognition system.So in order to improve the accuracy of segmentation traffic signs, this paper according to the unique nature of the traffic sign colors test of two kinds of segmentation method:(1) using the color threshold value method is proposed to realize the division of the traffic signs. Due to traffic signs in complex background, in order to improve the accuracy of traffic signs recognition, this paper first experiment in HSV, LAB,a single color space, select threshold to segment traffic signs; Then according tothe threshold of a single color space the value of the segmentation results and analysis of the data obtained HSV and LAB two color space threshold segmentation method of superposition, finally summarizes the experiment data, and concludes the advantages and disadvantages of the color threshold segmentation algorithm.(2) Traffic signs segmentation based on color template matching algorithm is the second method in this topic in the experiment.The specific work:first is the acquisition color sample, based on HSV, LAB, RG three chromaticity space structure of the two color template; Then according to the specific different matching algorithms will be two types of color templates respectively and experimental image matching division, and finally get the goal signs. The contrast matching effect chart to analyze the characteristics of this algorithm.The algorithm design of this article is in Matlab platform, statistics obtained experimental data of the various segmentation method to comparative analysis of experimental data,, finally choosing the optimal color segmentation algorithm to ensure that complex background of traffic signs right division.