动态图像序列中的运动目标检测

Detection of Moving Objects from Dynamic Image Sequence

根据动态图像序列中背景因成像过程中各种因素而产生变化所存在的复杂性 ,提出了自适应的前景目标检测方法。首先 ,建立图像每一像素点的高斯分布模型 ,并根据序列中的当前帧及历史帧信息自适应地调整模型的参数。然后 ,结合图像帧间的差分信息以及灰度分布的先验概率等因素将图像从空间域映射至统计域。最后 ,在统计域中对前景目标进行鲁棒分割。实验的结果反映了该方法的有效性。

The background of dynamic image sequence is very complex due to the factors in imaging process. To cope with this problem, an adaptive image segmentation method for detecting the foreground objects is proposed. First, a Gaussian distribution model for image pixel is proposed. The parameters contained in the model are adaptively updated based on the information from the current and historical frames. Then, every frame is mapped from spatial domain to statistical domain incorporating the factors such as the difference image from the consecutive frames and the prior distribution of a pixel density. Finally, the foreground objects are robustly segmented in the statistical domain. Experimental results show the feasibility of the proposed methods.