Real-time vehicle detection based on quasi-shot segmentation

In this paper, a Quasi-shot Segmentation Vehicle Detection (QSVD) method is proposed to improve the accuracy and robustness of traffic flow detection. The QSVD method is based on histogram intersection method to find the biggest difference between successive frames. Adjacent frames are considered in when vehicle is detected in current frame. Relying on the QSVD method, vehicle can be detected by signal peak and valley search and gradual transition shot detection problem can be resolved. The initial video is used to accomplish detection directly without image preprocessing. Experiments show that vehicles can be detected effectively by the QSVD method. The detection accuracy is remained at 93% or more.