License plate location based on projection method and genetic algorithm

Abstract

License plate recognition is an important part of the Intelligent Transportation Systems. The key technologies of the license plate recognition license are plate location, character segmentation and character recognition, which license plate location technology is the core technology of the three technologies. There are two problems of license plate location algorithm, one is complex background interference; another is conditions such as weather, light image quality and difficult to locate. In this paper, the license plate location method based on projection and genetic algorithm method combination for these two issues. Projection method has the features of simple, rapid and easy to implement. The genetic algorithm has efficient, parallel, adaptive characteristics. For the case of this article, we can quickly find the global optimum, and to combine these two methods. To a certain extent it is good solution to these two problems, the versatility of the algorithm greatly enhance the advantages into full play, the positioning accuracy is improved. © 2013 Springer-Verlag.