基于背景差分的动目标检测算法研究

提出一种新的基于背景差分的动目标检测算法。综合利用序列图像的时、空域信息提取目标。用中值法建立初始背景,并采用一种自适应的背景更新策略。实验结果表明,该算法简单,鲁棒性好,能够解决全局光照缓变和突变的问题,在复杂环境中也能准确地检测出运动目标。

Moving Object Detection Algorithm Based on Background Difference

A novel algorithm for the detection of moving objects is proposed.Temporal and spatial information are fused to extract the objects.The method of median filtering is used to build the initial background.Then a new self-adaptive background updating scheme is adopted.Simulation results show that the algorithm is simple and robust.It is able to deal with both gradual and sudden global illumination changes.And the presented method is also proved to perform correctly in the complex environment.