With the rapid progress of the technology of computer, image processing and analyzing technology has gained quick development. It has been applied to varied fields successfully and broadly. But it has some limitations in industrial practice because of no uniform resolution, so we can say that it is still a great challenge to solve practical problems with the image processing technology in the industry field. Now, the application research of image processing becomes one of important fields in science research area.The dissertation introduced the theory of image processing. The main work of it is as follows:First it discussed how to construct a vehicle plate recognition system with the image processing technology. Secondly, it analyzed the methods of image processing ,compared deeply with algorithms of the pretreatment of image and license plate locating related to LPRS; presented a license plate locating algorithm based on the vertical texture characteristics of characters, which can exactly and quickly locate the license plate ; Thirdly, it studied the standard algorithms of image processing such as image binaryzation ,edge detection, image segmentation ,image smoothing, and image enhancement. By combination of edge detection method and Hough changes, it effectively discriminated different type vehicles license plates, inclined and calculated some parameters using the characteristic projection method. Then ,it introduced the concept of character narrow region and applied this concept to the character region classified processing, enhanced the division accuracy; Finally, by using the method of feature matching and analyzing structure feature or sorts of statistics feature, it designed series classification to recognize the segment characters.This dissertation carried on a lot of experiments to validate the validity of the above algorithm using many kinds of car license pictures. The results of experiments showed that the methods are validity .These methods can not only enhanced car license processing accuracy, but also can reduce the car license process time in some degree. Therefore, these methods can be applied to the LPRS effectively. |