| 基于 Kalman滤波理论的运动目标检测新方法 |
| |
| 引用本文： | 任臣,张覃平.基于 Kalman滤波理论的运动目标检测新方法[J].光电工程,2007,34(4):7-11. |
| |
| 作者姓名： | 任臣 张覃平 |
| |
| 作者单位： | 1. 中国科学院光电技术研究所,四川,成都,610209;中国科学院研究生院,北京,100039
2. 中国科学院光电技术研究所,四川,成都,610209 |
| |
| 基金项目： | 国家高技术研究发展计划(863计划) |
| |
| 摘 要： | 该丈研究了基于Kalman滤波理论的渐消记忆最小二乘法在图像背景重建中的应用,并将它应用在复杂背景的图像序列中,实现对运动目标的自动检测.首先用渐消记忆最小二乘法对复杂背景进行预测和更新,然后把当前帧与预测的背景模板做差分运算,最后采用自适应阈值分割技术实现对目标的自动分割.文中通过对序列图像的仿真,讨论了最小二乘法的存在问题,改进及适用情况,干扰的消除.试验结果表明,该方法具有很强的实用性.
|
| 关 键 词： | Kalman滤波 背景预测 自适应阈值选择 运动目标检测 |
| 文章编号： | 1003-501X（2007）04-0007-05 |
| 收稿时间： | 2006/6/10 |
| 修稿时间： | 2006-06-102006-12-25 |
New method for detecting of moving targets based on Kalman filter theory |
| |
| REN Chen,ZHANG Qin-ping.New method for detecting of moving targets based on Kalman filter theory[J].Opto-Electronic Engineering,2007,34(4):7-11. |
| |
| Authors: | REN Chen ZHANG Qin-ping |
| |
| Affiliation: | 1. The Institute of Optics and Electronics, the Chinese Academy of Sciences, Chengdu 610209, China; 2. Graduate School of the Chinese Academy of Sciences, Beijing 100039, China |
| |
| Abstract: | The fading memory recursive least-squares based on Kalman filter is applied to rebuild image background. This method is used for detecting moving targets in the image sequence of clutter background. First, the rebuilt algorithm is adopted to predict and update the image background. And then a difference image can be obtained by subtracting the current frame from the predicted background. At last, the moving targets are automatically segmented by selecting an adaptive threshold. The problems which exist in the fading memory recursive least-squares, the improvement mad the elimination of disturbance are discussed by experiment. Experimental results are given to demonstrate the effectiveness and practicality of the method. |
| |
| Keywords: | Kalman filter Background prediction Adaptive threshold selection Moving targets detection |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
