摘要
运动图像序列分割是计算机视觉中的一个重要问题.本文采用基于贝叶斯框架的最大后验边缘概率算法进行运动目标分割.首先,重新定义贝叶斯框架中似然函数的平滑项,并采用区域收缩算法实现迭代过程中运动目标支持区的估计.然后提出一种通过区域中心和主轴表示6参数仿射运动的模型,通过区域主轴像素估计运动参数,提高算法执行速度,将估计问题转化为一个取值有界的最优化问题,采用 DIRECT 算法估计运动参数.该方法与传统方法相比,提高运动参数估计的准确性和稳定性.通过仿真实验结果证明该方法的有效性.
Segmentation of motion image sequences is an important problem in computer vision. In this paper, maximizer of the posterior marginals-maximum a posteriori (MPM-MAP), is adopted based on Bayesian frame for motion segmentation. Firstly, the smoothness term of likelihood function in Bayesian frame is redefined. The region shrinking algorithm is used to estimate the supporting regions of moving objects during the iteration. Then a model is proposed which represents the affine motion with 6 parameters by the center and main axes of a region. Motion parameters are estimated merely by pixels on main axes and derived more quickly than before. The estimation is transformed into a kind of optimal problem with parameters in limited ranges, and DIRECT algorithm is used to compute the motion parameters. Compared with the traditional algorithms, the proposed method improves the accuracy and stability in motion parameter estimation. The results of simulated experiments show the effectiveness of the proposed method.
出处
《模式识别与人工智能》
EI
CSCD
北大核心
2008年第2期214-220,共7页
Pattern Recognition and Artificial Intelligence
关键词
运动分割
最大后验边缘概率(MPM—MAP)
区域收缩算法
主轴仿射模型
DIRECT算法
Motion Segmentation, Maximizer of the Posterior Marginals-Maximum A Posteriori (MPM-MAP), Region Shrinking Algorithm, Affine Model of Main Axes, DIRECT Algorithm