Research Of Traffic Sign Recognition System Based On Support Vector Machine

ＡＢＳＴＲＡＣＴ　 　 With the development of society and economy, the jamming and huddling oftraffic become more and more serious, and become one of the bottle-neck of mordencity, traffic accidents are paid much attention by the governments of the world. In theconditions, ITS(Intelligent Traffic System) is come into being. It involves thetechnologies of pattern recognition, digital image processing, artificial intelligence,electronic technology, communication technology, system engineering and so no. Atpresent, many countries are busy in the researching and erecting of their own ITS. Moving vehicles collect the information of the circumference and give the hint oralert to the driver timely, a lot of traffic accidents can be avoid. It is the main work ofthe safety supplementary system. Traffic sign is the mark being fixed on the road orthe sides, it aims at hinting and altering the driver the road conditions ahead orrestricting some driving deeds. Consequently, traffic accidents are avoiding. So thesystem of automatic segment and recognition of traffic sign is the important softsupport of ITS and has great worth. Since the images being processed in traffic sign recognition system mostly havecomplex background, especially traffic sign have the problems of light, fading,distortion, defiling act. At present, the technologies of traffic sign recognition systembased on our country are immature, it is still limited in lab. In this article, supportvector machine is integrated to pattern recognition and a new way of traffic signrecognition is presented. The keystone and difficulty lies in the segment of traffic sign, feature abstractingand the design of classifier. In the research of traffic segment, through the analysis ofthe traffic sign characters, a segment way based on RGB model is proposed. The wayis free from lighting and the change of model is not necessary, so the speed of thealgorithm is improved; Because our final propose aims at the traffic sign recognitionsystem can meet the real-time need, we select the simple features that can representthe characters of traffic sign; Because the features of SVM and the potential in traffic III北京工业大学工学硕士学位论文sign recognition, we bring forward a hierarchical recognition modes based on SVMafter plenty of researching. The model is divided into three steps and fulfill therecognition task of sign and not sign, type of sign, specific sign orderly.