Vehicle video detection based on pulsed coupled neural network

Vehicle Video detection is an important part of intelligent transportation system for its characteristics of effectiveness and accuracy, and it provides important data for the security of urban traffic operation. However, in order to improve the reliability of vehicle detection in the traffic video detection system, in this paper, a pulse-coupled neural network (PCNN) is used for the image segmentation to obtain the target vehicle, automatic propagation characteristic of the pulse-coupled neural network is utilized to perform the morphological restoration of the vehicles, which is resulted from the consideration of the impact from the external environment and the interference in the transmission of data as well as the real-time requirements of video detection. The experimental results show that a good effect is obtained in video detection by using the proposed method.