A Novel Vehicle Flow Detection Algorithm Based on Motion Saliency for Traffic Surveillance System

Abstract

Traffic Flow Detection plays an important role in the field of Intelligent Transportation Systems (ITS). Traffic flow detection focuses on the detection and segmentation of video object. The most existing methods need to implement complex background modeling, and accordingly increase the computing complexity and computing cost. In order to reduce the computing cost of the vehicle detection, we propose a new vehicle detection method based on saliency energy image (SEI) and saliency motion energy image (SMEI) for automatic traffic flow detection. First, we set the detecting region of objects, and computing image saliency map of the detecting region for each frame. Then saliency energy image (SEI) and saliency motion energy image (SMEI) are calculated. Finally, the vehicle flow is detected by combining the vertical projection histogram of the SEIs and the binary SMEIs within pre-set virtual detecting box. Experimental results show that our method can work in real-time with a high accuracy and robustness to noise.