Highway Toll System, The Key Technology Of Image Recognition

The highway toll system is an important component of the highway management system and Intelligent Transportation System. With the development of the highway network and the increment of the automotives, the highway toll system faces great challenge. Image Processing Technology is introduced in this paper to improve the efficiency of the toll system and prevent any cheat. Key technologies such as Image Processing and Pattern Recognition are intensively studied in this paper. Such research would be very valuable both in academic and practical areas.Number Plate Recognition is mainly studied in this paper, which includes Image Acquisition, Number Plate Location, Character Segmentation and Character Recognition. The details are as follow:1. Image Acquisition: The Acquisition devices are carefully selected by comparison. A series of solution measurements are proposed according to different practical conditions.2. Number Plate Location: A location scheme based on the texture features of characters is proposed.3. Character Segmentation: Firstly, the number plate is horizontally segmented by multi-angle parallel scanning. Then the vertical segmentation is carried out by an adaptable template.4. Character Recognition: Two classifiers are respectively designed for Chinese characters and English characters and numbers, among which, a recognition method based on the structure features for English characters and numbers is proposed in this paper.By improvement and optimization in practice, the experimental results are satisfactory. Further more, Several Automotive Style Recognition methods are primarily studied, such as fuzzy method, neural network and the templatematching method according to three-dimension automotive style.A Number Plate Recognition System based on the key technologies in thispaper has been adopted by some highways.