一种基于三帧差分与高斯模型的运动目标检测算法

his paper proposes a motion objects detection algorithm based on three frame differencing and Gaussian model. Firstly, a reliable Gaussian background of each pixel is established on three frame differencing images, and to adapt to the complicated variety of the scene using Gaussian background model updating mechanism. Then, two frames are done logic \"and\" operation after matching Gaussian background model. Finally, post-processing must be done as necessary to get accurate motion objects. In the software development platform of Matlab7.1, Comparative experiments on three frames differencing method, Gaussian Mixture Model background differencing method and the algorithm proposed by this paper. The simulation results show that this algorithm runs simply, can detect effectively the motion objects, and fits for real time detection.