Research On Character Segmentation Methods For Vehicle License Plate

License plate recognition system is an important subject to be studied in the intelligent traffic. It is mainly composed of three key parts: license plate location, license plate character segmentation and license plate character recognition. License plate character segmentation , as a key part of the license plate recognition system , is mainly studied in this paper.Firstly, the preprocessing algorithm for license plate character segmentation is thoroughly studied, especially binary conversion. According to the color features of license plate image, it is presented in this paper that the gray adaptive binarization is combined with the colorful image binarization based on neural network, and a better effect of binarization is obtained.Secondly, according to the connectivity of numbers and letters on the license plate, it is presented that the characters are segmented through the combination of the mathematical morphology and the projection. The experimental results show that it is effective. Lastly, according to the structure feature and pixel distribution of the license plates, an approach for license plate character segmentation based on pixel distribution is presented.The experimental results show that it is effective, and the segmentation is still clear, especially for the situation where character-touching and noise exsist among license plate.