摘要
提出一种基于动态阈值对称差分和背景差法的运动对象检测算法。首先通过建立一个基于统计的可靠背景更新模型,由背景差法得到基本准确的前景图像;然后与用对称差分法得到的差分图像综合;最后得到完整可靠的运动目标图像。中间采用了一种动态的最优阈值获取方法,然后用形态学滤波和连通区域面积检测进行后处理,以消除噪声和背景扰动带来的影响,并用区域填充算法来填补目标区域的小孔,从而将视频序列中的运动目标比较可靠地检测出来。实验结果表明,该方法快速、准确,有一定的实际应用价值。
A novel algorithm for moving detection, which employed adaptive background subtraction and symmetrical differencing method, was presented. A modified selective updating model was proposed as the reliable adaptive statistical background updating method, and the background subtraction was combined with symmetrical differencing to detect moving information. This paper also presented a dynamic optimization threshold method for image. After the motion detection operation, morphologic filtering and connected region area measurement were introduced to suppress the noise and solve the background disturbance problem. Then the area filling algorithm was used to fill the small hole in the detected object, Finally the moving objects were extracted reliably, The experiment result shows that the presented algorithm run quickly and veraciously.
出处
《计算机应用研究》
CSCD
北大核心
2008年第2期488-490,494,共4页
Application Research of Computers
基金
国家自然科学基金资助项目(60573079,60673093)
关键词
背景差
对称差分
背景模型
运动检测
background subtraction
symmetrical differencing
background model
moving object detection
