Automatic Estimation of Crowd Density in Complex Scene
WU Ximei · Modern Electronics Technique · 2009
Crowd density estimation,which means a lot to the crowd safety and timely management and scheduling,is an essential issue in scene surveillance.As the existed method for crowd density estimation works only in a simple background scene,an improved method which can apply to complicated scene is proposed in this paper.First background is updated by adaptive background estimation,then employs mask and background subtracting to get the crowd foreground,the crowd features are extracted by employing grey level dependence matrix,at last uses the support vector machine to identify the level of the crowd density.The experimental results show that the method is efficient and feasible,which can apply to complicated scene.