License Plate Recognition Technology Research

With the rapid development of economy in China, more and more expressways, urban roads and parking lots have been constructed, which requires higher traffic control and safety management. Intelligent transportation system has become an important trend of current traffic control management, and its wide application will promote the traffic management more intelligent, automatic and standardized.License plate recognition system is mainly composed of four parts:license plate image acquisition, locating, character segmentation and character recognition. This paper briefly introduces the composition of license plate recognition system, its technology development history and application status. The license plate character recognition is extensively studied. The main contents are listed as follows:1、Based on the analysis of two kinds of character segmentation algorithms, a binarization threshold of license plate character segmentation algorithm is derived. This study adds morphology of corrosion and expansion to the ordinary binarization algorithm. Thus, the problem can be better solved when the plate comes with objective factors such as light and motion, and thus obtains the good effect of binarization.2、Due to real-time requirement of our system, this paper proposes the software experiment solution combing license plate recognition technology and embedded system, with the embedded Linux system as the test platform and GNU open source development tools as the debugging tool. This paper also concludes the advantages of embedded license plate recognition system.