Research On License Plate Automatic Recognition Method

At present, License Plate Automatic Recognition is the important core of intelligent transportation system, an important part of public service traffic information management system, the foundation of normal operation of the intelligent transportation system.The paper studies the use of mathematical morphology by combining with other methods to improve the effect of the license plate location and character segmentation, and further enhance the efficiency of license plate recognition. We use the method of combining mathematical morphology and topology to locate the plate region extraction, use the connectivity posture analysis method of rough positioning of the plate image to deal with the license number of the candidate Euler discrimination and extraction of real plate region. We use the method of combining mathematical morphology and projection to segment the license plate character, the extraction of characters using the template matching method for vehicle license character recognition. In the paper, we use Matlab software to achieve the algorithm about License Plate Automatic Recognition. The experiment results show that the use of mathematical morphology method license plate localization accuracy is higher, character segmentation quality is good, the identification accuracy is higher.