Research On License Plate Location Based On Digital Image Processing

With the rapid development of our national economy, Intelligent Transportation Systems will become the trend of modern traffic management development. Automatic vehicle identification is a premise for Intelligent Transportation System (ITS). License Plate Recognition (LPR) is the most promising technique in the automatic vehicle identification system.LPR consists of License Plate Location, Character Segmentation and Character Recognition. License Plate Location is the base of Character Segmentation and Character Recognition, plays a crucial role in the performance of whole system, and determines the speed and accuracy of system identification.Seen the nature of image processing, License Plate Location includes the License Plate Location based on gray image and License Plate Location based on color image. License Plate Location based on gray image is faster, whose accuracy is not high and affected by the environment, light and other factors. Color image's information is much more than gray image. And is easier posited. But the amount of computation is larger. Therefore, the research of license plate location Based on color image is smaller. In recent years, with the rapid development of computer technology and image processing, License Plate Location based on color image will be a hot spot.This paper researched on the Location of color automobile license plate image by digital image processing techniques. In this paper, firstly we analyzed the difficulties and development status of license plate positioning technology. Then we introduced in detail the basic concepts, mathematics, basic operations, research, processing methods and application areas of digital image processing technology. Next we researched on the existing license plate positioning technology, and introduced an approach to license plate location based on pixels classification. According to the differences in the RGB value of the license plate color to classify the pixels, we analyze and statistics pixels, and take advantage of license plate form character and color character to determine license plate position. Experimental results show that the method the approach is fast and accurate, and applies to license plate location in cases of various backgrounds and positions.