Research On License Plate Location And Character Segmentation Based On Gray Image

With the development of modern traffic, Intelligent Transportation System has been the main developing direction in the management of the traffic. As the core of the ITS, License Plate Recognition plays an important role now. In the full understanding and grasp of the current status of research of LPR system , papers make deep study on license plate location and character segmentation with the Chinese situation .Through deeply research in license plate character and license plate location, a license plate location algorithm based on horizontal scan and vertical projection is put forward that can search the license plate area accurately; after license plate location, character segmentation is implemented by 4 steps: skew correction of the license region, edge examination and image binaryzation, level boundary resection,and vertical segmentation of characters. In the step of edge examination and image binaryzation a novel character extraction algorithm is presented, this algorithm employs Canny operator to extract the characters in gray license plate image. After extracting the characters, this algorithm adopts the iterative segmentation algorithm to gain the best threshold and process the image, and then combines the morphology algorithms to fill the gaps in the characters. Canny operator can extract the edge of binary image more completely, and it is advantageous in vertical projection after filling the gaps of the characters.The algorithm of this paper is implemented in the programming environment of Visual C++ 6.0. The experimental results indicate that compared with other traditional algorithms, it not only can effectively extract the characters and evidently filter noises, but also protect the edges of characters. Moreover, it can provide the techno-guarantees for the following license plate recognition.