Detection And Recognition Algorithm Of Traffic Signs Under Complex Condition

In recent years, computer technology as the representative of the information technology industry has been rapid development of high-end countries in the field of information into a growing number of human and material resources to support the development of high-end technology industries.Among them, with the increase of vehicles on the road traffic congestion and its adverse effects on the environment so that more and more scientists began to research and research unmanned vehicle technology.The core of the study covers two main functions, one is based on computer-aided visual perception of the surrounding traffic scene and traffic conditions to determine the other is the process of moving vehicles around the corresponding traffic signs of the automatic identification and detection. The above scientific research can provide the necessary theoretical support and decision-making basis for the real meaning of vehicles to realize "unmanned" driving. The main work of this paper is to integrate traffic detection and recognition algorithm based on traffic signs at home and abroad, and to research, As well as automatic detection of information and systems. At the same time, this topic is also based on the theoretical derivation, the practical application of research. The detection and recognition system of traffic signs in complex environment is realized.The innovation of this subject mainly includes the following aspects:(1) Realization of more than 30 kinds of traffic signs based on symbol type detection and recognition algorithm;(2) After transforming the original RGB space image, the original characteristic color of the traffic sign is highlighted, and the threshold segmentation image is re-selected.(3) In order to reduce the false detection rate, this subject re-processing of the morphological data processing results again for this purpose to build a new effective perception of the traffic signs of the edge of the region.(4) On the basis of the above-mentioned shape of the edge area, we can exclude other interference factors in the feature setting and screening of the shape "marker". The experimental results show that the traffic signs can be detected and extracted more accurately in complex urban traffic scenarios, and the identification of the extracted traffic signs can be identified and the accuracy of identification is high. The traffic sign detection and recognition algorithm realized in this project still has Better performance.In order to construct the multi-target tracking model of traffic signs, the paper defines the status of traffic marking targets. The Kalman filter algorithm based on the trackless Kalman filter is implemented, and the model is constructed on the basis of the model. The model is built for the trajectory and state of a single traffic sign, and finally the determination of the expected target position is achieved.In this paper, through the study of traffic sign detection and recognition algorithm, a simulation program for automatic detection and recognition of traffic signs under complex conditions is realized by experiments. The program includes the detection and recognition module of traffic signs, Can mark, extract and display traffic signs. It can provide unmanned vehicles with driving information content for road environment under complicated conditions, and can also provide reference for other related research.