Location And Recognition Of License Plate Based On Mathematical Morphology

License Plate Recognition System (LPRS)-to carry out the traffic management, which is an important subject of these fields in Computer Vision, Pattern Recognition and Image Processing technologies. It has special importance or significance in theory and practical applications.This paper studies in depth the theoretical method of the Mathematical Morphology, the License Plate Recognition, the Character Segmentation and the Character Recognition technologies, and summaries the applications of adopting the Mathematical-Morphology-based processing technologies in the LPRS, as well as applying the BP Neural Network in Character Recognition. It also analyses the reason why the efficiency of the LPRS is so low, and proposes the methods of the morphology-based Image Processing and the cluster-based method of Character Segmentation with reference to the projection method.On the base of the design and function zoning of the morphology-based License Plate Location and Recognition, this paper discusses in details about the process of License Plate Location and Character Segmentation, and discuss the neural network and related theories, which focuses on analysis of the neural networks in Character Recognition application and the design of BP neural network classifiers and extraction of the license Plate characters. All the characteristics of the char image are input to BP network and the output of BP network is coded by binary code. It provides the procedures of processing and computing through the extraction of original images and the pre-process of images, as well as the morphology-based method. This paper shows the results of the MATLAB simulation experiment. It is proved through the analysis of the experiment results that, it becomes much easier in computing, more efficient in filtering the image noises and also offers great advantages in Character Segmentation by adopting the Mathematical-Morphology-based technologies of processing the license plate.Finally, this paper summarizes and prospects the Follow-up work and algorithms.