FPGA based real-time vehicle detection system under complex background

In this paper, a real-time vehicle detection system is designed and implemented on an FPGA (Field Programmable Gate Array). The system is composed of an infrared camera and an image acquisition and processing board developed by our research team. An FPGA chip and a DSP chip are embedded in the image board as the major calculation units, which make realtime computation possible. First, edge features of the training samples are extracted according to Naive Bayes Model and then fused with brightness features to construct the detector template on the host computer. Second, real-time vehicle detection experiments are conducted according to the detector template on the image acquisition and processing board. In contrast to conventional methods in which a huge number of negative samples are collected for training, an even distribution is assumed in our method and no negative samples are needed, thus greatly shortening the training time. Experiment results show that the algorithm can detect different shapes and sizes of vehicle objects efficiently.