License Plate Recognition Technology Application

The License Plate Identification System is an important component of Intelligent Transportation System, and can be applied in many fields. Its implementation mainly depends on the computer vision technology and the pattern recognition technology. A typical LPR System consists of a video image-acquisition subsystem, an image-preprocessing unit, a plate character isolation and recognition engine.My work including: license plate locating, character segmentation and character recognition .The related system implementing above function is completed.License plate locating is a sticking point in my work. Its result has an important impact on the accuracy of the final result. Algorithms based on license plate texture and gray level character is applied in license plate locating. The method to eliminate noise through binarization and mathematical morphologic. Fuzzy pattern recognition algorithms is applied to perfect the locating. The paper introduce all this and the related data illuminate the efficiency and veracity of the new method.Character segmentation consists of rim elimination and character isolation. Algorithms based on edge detection of license plate character prior knowledge and gray level projection is applied. And eliminating the noise of the segmentation result effectively.Character recognition is based on modified pattern matching and morphology thinning. In my work pruning of the thinning character is applied to improve the elegance of the character. The ingenious pattern is used to improve the efficiency of recognition. They all are recommended in my paper.