Study On Vehicle Logo Detection & Recognition Based On Video Image

Recognition system of vehicle mainly consists of two key technologies: recognition of vehicle_license_plate (VLP) and recognition of vehicle type. It not only finds wide application in ITS, but also is a hot point of research in computer vision, image processing and pattern recognition. So its related technology is attended prevalently. In our paper, we pay attention to the location and recognition of vehicle logo. The methods to deal with above problems are proposed in the paper and testified in the experiment。The main content and innovation of the dissertation are as follows:(1) . Vehicle-logo location algorithm: While VLP has evident texture feature and regular shape, the texture feature, shape and size of every vehicle-logo are different. It can consume much time if we adopt the general object location algorithm, such asrelation match. So the problem of vehicle-logo location has been a bottleneck in the real-time application of vehicle-logo recognition. For the sake of such solutions, a fast and robust method of vehicle-logo location is presented in this paper. By full utilize of the location of vehicle-license-plate, and its correlation of vehicle-logo and vehicle-head, and by utilize of axis symmetry of vehicle-head, an experimental rough searching rectangle of vehicle-logo can be located. Further on, by an adaptive morphology filter and an improved template matching method, the vehicle-logo can be accurately located.(2). Vehicle-logo recognition algorithm: After location, vehiclelogo recognition becomes a 2-D shape recognition problem. But being subject to illumination, noise, shape similarity and part occlusion, it is hard to get satisfactory results by common methods. To improve the accurate of recognition, a vehicle logo recognition method which integrates template matching with feature matching is proposed. Experimental results approve the accurate of the method.The topic based on the computer vision, pattern recognition and image processing technology, the key problems, VLP vehicle-type (especially vehicle-logo) recognition are studied thoroughly in the dissertation. A set of practical methods is proposed to deal with the above problems, and also to resolve common problems in ATR theoretically. Experimental results show that these methods are reasonable and have large theoretical and practical values.