为实现教室监控视频中多目标智能跟踪，研究提出了一种基于前景检测修正和梯度特征的改进型ＣａｍＳｈｉｆｔ算法。该算法利用收敛速度分段分区可调的背景重建算法对视频序列进行背景重建与前景检测。综合考虑了前景检测结果，图像

梯度和颜色特征，进行ＣａｍＳｈｉｆｔ目标跟踪。该算法能自适应调节梯度与颜色特征融合比例，对强光有一定的抑制作用。开发了实验软件，仿真结果表明了所提算法的有效性，为解决教室照明用电浪费问题提供一个可行的解决方案。-Surveillance video in the classroom to achieve multi-target intelligent tracking study raises the prospect of improved CamShift algorithm based on the detection correction and gradient features. The algorithm uses the convergence speed segmented partitions adjustable background reconstruction algorithm for video sequences background and foreground detection reconstruction. Considering the prospects of the test results, the image gradient and color features, make CamShift target tracking. The algorithm can adaptively adjust the gradient and color feature fusion proportion of light has a certain extent. Developed the experimental software, simulation results show the effectiveness of the proposed algorithm for solving the waste problem classroom lighting electricity to provide a viable solution.