基于增强的粒子滤波算法的医学图像动态轮廓跟踪新方法

南方医科大学学报 ›› 2004, Vol. 24 ›› Issue (06): 677-681.

基于增强的粒子滤波算法的医学图像动态轮廓跟踪新方法

周寿军, 陈武凡

第一军医大学医学图像全军重点实验室, 广东, 广州, 510515

出版日期:2004-06-20 发布日期:2004-06-20
基金资助:
收稿日期:2003-9-16。
基金项目:国家自然科学青年科学基金项目(60302022);国家自然科学重点项目(30130180)
作者简介:周寿军(1971- ),男,第一军医大学在读博士研究生,主要研究方向为数字图像压缩与图像传输.
通讯作者:陈武凡(1949- ),男,第一军医大学医学图像全军重点实验室主任、教授、博士生导师,主要研究领域为数字图像处理与模式识别.

Dynamic contour tracking of medical images based on improved particle filter

ZHOU Shou-jun, CHEN Wu-fan

第一军医大学医学图像全军重点实验室, 广东, 广州, 510515

Online:2004-06-20 Published:2004-06-20

摘要/Abstract

摘要： 关于医学图像的研究,感兴趣区的运动估计和跟踪是一个深受关注的领域。鉴于医学图像质量低、噪声大的普遍特点,从状态变量的非线性、非高斯分布前提出发,利用粒子滤波技术解决该类跟踪问题是一种具有挑战性的技术:由于经典粒子滤波器的权值计算,尤其是重要密度函数的构造方法严重影响了粒子滤波器的性能,本文提出了重要改进。针对用粒子滤波方法估计动态轮廓线这一特殊应用,构造了具有特色的似然和先验概率密度算法。结合客观的理论评价标准和大量比较试验,该方法为精确估计动态轮廓线提供了较好的解决对策。

Abstract: In the research of medical image processing, motion estimation and tracking relating to the region of interest has been given considerable attention. For improving the quality of the noisy or cluttered medical images, the particle filter (PF) based on the non-linear and non-Gaussian Bayesian State Estimation is a better as well as a technically challenging solution. As the algorithm of particle weights, especially the importance density function, often severely affects the performance of the PF, we propose in this paper a better algorithm for its improvement; in addition, to ensure better tracking of the dynamic contour with the PF, we proposed a new algorithm for the likelihood and prior probability density. Objective theoretical evaluation and substantial comparative experiments suggest that this method can be a good solution for accurate dynamic contour tracking.

中图分类号:

TP391

周寿军, 陈武凡. 基于增强的粒子滤波算法的医学图像动态轮廓跟踪新方法[J]. 南方医科大学学报, 2004, 24(06): 677-681.

ZHOU Shou-jun, CHEN Wu-fan. Dynamic contour tracking of medical images based on improved particle filter[J]. Journal of Southern Medical University, 2004, 24(06): 677-681.

参考文献

[1] Bjorn S,Arasanathan T,Torr PH,et al.Filtering using a tree-based estimator [C].Proc.9th Int.Conf.on Computer Vision,France:Nice,2003.10.
[2] Arasanathan T,Bjorn S,Tort PH,et al.Learning a kinem-atic prior for tree-based filtering [C].Proc British Machine Vision Conference,UK:Norwich,2003.9.
[3] Fabler R,Bouthemy P.Non parametric motion recognition using temporal multiscale Gibbs models [C].IEEE Confon Computer Vision and Pattern Recognition,Hawai,CVPR’01,2001.12.
[4] Robin DM.Image sequence restoration using Gibbs distributions [D].A thesis submitted to the University of Cambridge for the degree of Doctor of Philosophy,Department of Engineering,1995.
[5] van-Merwe R,Doucet A.The unscented particle filter[R].Technical Report CUED/F-INFENG/TR 380.Department of Engineering,Cambridge University,2000.
[6] Hisashi T.Nonlinear and Non-Ganssian state-space modeling with Monte Carlo techniques:A survey and comparative study [C].Faculty of Economics,Kobe University,Japan:Kobe,2000.1.
[7] Sanjeev A,Simon M,Neil G,et al.A tutorial on particle fitlers for on-line non-linear/non-Ganssian Bayesian tracking [J].IEEE Trans Sign Proc,2002,50(2):174-88.
[8] LiuJS,Chen R.Sequential Monte Carlo methods for dynamical systems[J].JAm Stat Assoc,1988,93:1032-44.
[9] Doucet A,Gordon NJ.Simulation-based optical filter for manoeuvring target tracking [C].SP1E Signal and Data Processing of Small Targets,1999.3809.
[10] Bergrnan N.Recursive Bayesian estimation:Navigation and tracking application [D].Linkoping University,1999.
[11] Xu CY,Jerry LP.Snakes,shapes,and gradient vector flow[J].IEEE Trans lmag Proc,1998,7(3):359-69.
[12] Zahn CT,Roskies RZ.Fourier descriptors for plane closed curves[J].IEEE Trans Comp,1972,21:269-81.
[13] Matthew O,Willian F.A Bayesian approach to tracking multiple targets using sensor arrays and particle fitlers [D].Department of Engineering,Cambridge University,2001.