Vehicle recognition based on saliency detection and color histogram

Vehicle recognition is one of the important parts in the intelligent transportation system. A fast vehicle recognition method is proposed to improve the performance of the vehicle recognition. Firstly, the detected image is preprocessed by the image saliency detection based on local features, which can eliminate a large number of background information to highlight the vehicle image. Then, the vehicle image which is converted color space from RGB to HSV, and the H component is counted by color histogram which is used to identify the vehicle. Finally, the positions of the vehicles are determined. The simulation results show that the method can detect the vehicles from the complex background.