The Car License Plate Recognition Technology And System Development

Automatic vehicle identification plays an important role in the intelligent control and management of the modern traffic. As the key technology of automatic vehicle identification, the license plate recognition(LPR) has attracted more and more attention.In this paper, the key technologies of LPR are researched and discussed. After that the LPR prototype system is developed.The vehicle image processing is the first step of LPR. After the experiments of various vehicle images, it has been found out that Binsharpen-Midvalue Filter method is more appropriate. Through the processing, the noise on the image is eliminated effectively and the vertical texture is clearer.Location of the license plate is the key step in the LPR technology. Two methods for the location of license plate are proposed based on distinguished texture feature. One is window search and shift scan, the other is fractal box dimension inspection. The experimental results show that the two methods are excellent in the accuracy and the robustness.Character recognition is another key step in the LPR technology. How to correct the slant plate and segment character is discussed. The method based on character strokes for the slant plate correction and the projection-histogram method for character segmentation are proposed. Several simple recognizers are combined together to recognize characters by utilizing the peculiarity of each one. The experiments show all these methods are pretty good and the proper license plate locating rate is about 98% and the character recognition accuracy reaches 89%.In the end, the LPR prototype system is developed on Visual Basic and Windows. It is not only friendly but also practical.