Intelligent transportation is the future direction of traffic, and the video-based vehicle detection and tracking is one of the key technologies. Traffic condition is affected by climate, light and other factors, its changes are very complex so as to affect the effect of vehicle detection. The paper puts forward a background reconstruction algorithm of dynamic updating. Through comparing the current frame with background frame, the method obtains the dynamic frame of continuous changes to build new background image, and then to achieve dynamic-updating background. And according to the background obtained, the method detects moving targets with background subtraction and it will be applied to video-vehicle detection. Experimental results show that the algorithm is very fast for reconstruction and has achieved good results in the video vehicle detection.
