This paper presents the main modules of the system to efficiently recognize type of the vehicles. The entry results are in line with our expectations. The proposed system achieves good performances on a test set containing over 3500 vehicle images and the detection rate is about 93% when it was compared with the measurements done by a human expert. Moreover, it is not sensitive to variation in time, weather and light condition. The computational complexity is low and the algorithm can work in real time.
展开▼