License Plate Character Segmentation And Recognition Algorithm Research

License Plate Recognition (LPR) system is one of the most important parts in Intelligent Transport System (ITS), the characters segmentation and the characters recognitiont are the key technology of the LPR system.In this paper, a characters segmentation algorithm based on types of past algorithm is put forward. which is fast and its accuracy is high. The key technologies of LPR are researched and discussed in this algorithm, includeing gray Preprocess of image, image binarization, gradient emendition, The Horizontal of segmentation and vertical segmentation, character normalization。Character segmentation algorithm in which the most critical part is the vertical segmentation algorithm which use of characters in a fixed aspect ratio features, extract the character height, and get the character width, segment the license plate character. In addition, this paper has done improved research about the vertical traverse density (VTD) and horizontal traverse density (HTD)character recognition algorithm, through pre-processing, process control, the second classification, etc., identify improvements, the recognition algorithm of the recognition accuracy has been great increase.Finally, by choosing pictures to test different scenarios, character segmentation algorithm. This algorithm is simple, fast, high accuracy license plate recognition system to meet the real-time performance, versatility, and accuracy requirements, has a certain value.