The Design And Application Of Automatic License Plate Recognition System Based On Improved Neural Network

With the fantastic spur in economy and rapid development of the owning amount of automobile, the highway communication becomes one of the most important communications and transportation ways in our country, but the traditional transportation can't meet practical demands nowadays, also crowded traffic needs more advanced and more effective traffic administration and system. So ITS (Intelligent Transportation Systems) which uses image preprocessing technology and communication technology becomes main research direction. Based on this background, the algorithm that recognizes the license plate and the license number automatically is designed using the visual analysis and the image processing technology.In this paper, the LPS (Automatic License Plate Identification System) is researched, solution is proposed and experiment has been finished. Based on this a set of software system has been developed. The architecture of system is introduced and the modules composed of the system are deeply researched. The technology of video gathering and image transitions are talked about. Algorithm with how to judge the vehicles violating regulations is put forward based on virtual coil and vehicle move track. In the modules of plate location and character isolation, the binary algorithm-Otsu is researched in-depth and the vertical projection to locate plate is used. Image rotation technology is used to complete tilt image correction and character is isolated with a simple and effective algorithm.In the part of plate recognition, neural network technology is used. Defects are improved that BP neural network convergences traditional slow and easy dives into local minimum points. Improved neural network is proposed to recognize license character. The experiment proved the algorithm is effective to plate recognition.