| In recent years, a sharp increase in the number of vehicle is emerging in our country.It is very important of how to effectively carry out related management such as vehicle running, vehicle identification, vehicle monitoring and so on. License plate recognition technology is an effective means of solving these problems. Currently, the license plate recognition technology has been widely applied in many fields such as highway fees, parking management, urban road traffic management, etc, which is obtained much attention of researchers. License plate recognition technology is the basis for intelligent traffic management systems and it is also the core technology. In generally, it mainly involves related knowledge and technology of digital image processing, signal processing and pattern recognition and other related disciplines.This paper studies the relevant principles and knowledge of license plate recognition technology, especially the following four key parts:pretreatment of the license plate, license plate location, character segmentation, character recognition. The license plate recognition system is achieved through C#combined with on this basis.100license plate images are used as an experimental, identification rate of91%with better recognition effect. The following four aspects are mainly studied in this paper.1) Pretreatment identifying images containing license plate using histogram equalization and median filtering method, which effectively improve contrast and sharpness of image, and eliminate image noise.2) Effectively locate license plate based on gray-hopping method combination with edge detection.3) In the processing of license plate character segmentation, the method of vertical projection and artificial correction are used in order to improve segmentation correct rate.4) Make binarization and normalized processing of divided characters.And then realize character recognition based on template matching algorithm by matching all the templates finally. |
PDF Full Text Request