A Mobile Electronic Police System based on Improve-ART1 Neural Network

This paper presents a mobile electronic police system whose core is an automatic license plate recognition system based on improve-ART1 neural network. Because of the characteristics of Chinese license plate, the research of automatic license plate recognition technology based on neural network focuses on the recognition accuracy, recognition speed, eases of training and learning and convergence. Experiments show the moving dynamic actual recognition rate of the system reached 98% when the vehicle speed is less than 140 km/h. Furthermore, the system was stable and reached the practical level when test targets and testers can have relative-motion.