一种最大后验概率条件下的运动目标检测方法

An Approach for Moving Object Detection Using Maximum a Posteriori

提出一种最大后验概率条件下的运动目标检测方法.首先根据条件随机场模型和马尔可夫随机场模型建立了一个最大后验概率框架.在该框架内融入了连续标记场的时域信息、颜色信息和每个标记场的空域信息.考虑到传统方法融入的特征信息不够,提取目标的准确度不高,在目标模型中充分融入了颜色信息和边缘特征,以便获得更好的检测效果.实验结果表明提出的方法能正确检测到运动目标.

An approach based on maximum a posteriori is presented for moving object detection in complex video scenes. Firstly, a maximum a posteriori framework is created according to conditional random field model and Markov random field model. Then temporal dependencies of consecutive label fields and spatial dependencies within each label field are merged into this framework. The object detection method integrates both color and edge features by object probability model. Experimental resuits show that the presented approach can accurately detect moving objects.