摘要
在视频的运动目标检测中,有很多经典方法,但是大部分都不能完整地从复杂背景中检测出运动目标。提出了一种运动目标检测的新算法——协方差矩阵算法,它集成了像素级和区域级2种不同的层次。在像素级中,空间的属性主要来之于像素的坐标值、亮度、纹理和梯度等。在区域级中,通过计算一个像素周围矩形区域的协方差矩阵来表示像素间的特征相关性。实验结果表明,该方法具有较强的鲁棒性,明显优于传统的目标检测算法。
Many methods can be used for moving object detection in video,but most of them can't detect moving object from complex background. The paper presents a method for moving object detection. The eovariance matrix algorithm integrates two different levels:pixel level and region level. At the pixel level, spatial properties that are obtained from pixel level coordinate value,intensity, texture,gradient and so on. At the region level,the correlation of pixels is represented by a covariance matrix that is calculated over a rectangle region around the pixel. The Experiment results show this method is more robust, and dramatically outperforms traditional method for moving object detection.
出处
《无线电通信技术》
2012年第4期51-53,59,共4页
Radio Communications Technology
关键词
目标检测
协方差矩阵
像素
区域
object detection
covariance matrix
pixel
region