基于状态分割的运动目标实时跟踪

针对基于云台的移动式摄像头视频监控系统，为准确、实时地对运动目标实施检测、跟踪，提出了一种基于状态分割思想的运动目标实时跟踪方法。该方法将运动目标检测跟踪过程按摄像头的运动状态分为静止、运动2个阶段。在摄像头静止阶段，采用基于混合高斯模型的背景差法检测运动目标，提取目标的颜色特征信息；在摄像头运动阶段，采用Camshift算法对运动目标进行跟踪。开发了基于 OpenCV 开源库的算法程序。实验结果表明，在目标颜色特征显著的情况下，该方法实现了移动式摄像头对运动目标的精确跟踪，并具有较好的鲁棒性和实时性。

Real Time Tracking for Moving Objects Based on State Segmentation Method

In order to optimize the video surveillance system for detecting and tracking the moving object accurately, a real-time tracking method for moving object based on state segmentation is proposed. This method divides the tracking process into two phases, namely, the static and motion phases, according to the state of the camera. In the static phase of the camera, the background subtraction based on Gaussian mixture model is used to detect the moving object, and the color features of the object is extracted.?In the camera motion stage,?the Camshift algorithm is applied to track the moving object. An algorithm program based on the OpenCV open source library is also developed in this study.?The experiment shows that the proposed method has good robustness and can accurately track the moving object with significant color features in real-time