摘要
目前报道的图像局部匹配方法一般假设参数变化符合线性或高斯分布,求解容易陷入局部极值,不能适应变形程度较大的情况。本文针对此问题提出了一种基于贝叶斯估计的方法,通过马尔科夫链蒙特卡罗(MCMC)算法求解后验概率分布的极值获得最优匹配参数。试验表明本文方法可以精确地匹配存在较大形变的点局部区域,并且可以成功地处理视频跟踪目标的尺度和方向变化。
As the reported methods are on the assumption that transformational parameters changes are followed by linear or Gaussian distribution so that the solutions cannot suit for large image deformations and are easy to get into local extremum. To solve this problem, a Bayesian estimation of point local patch registration parameters is presented in this paper. This method apply MCMC algorithm to solve the posterior probability distribution problem and obtain the optimal parameters. The test results prove this method can perform point local patch registration under large image deformations accurately and successfully coped with target scale or orientation variations in video tracking.
出处
《微计算机信息》
2010年第21期4-5,8,共3页
Control & Automation
基金
基金申请人:杨鸿波
项目名称:辐射图像分析关键技术的研究
基金颁发部门:北京市教育委员会(KM200810772001)
关键词
点局部匹配
贝叶斯估计
色彩分布
point local patch registration
Bayesian estimation
color distribution
