摘要
提出基于改进的弹簧质子模型的图像配准算法.通过将基于模板图像、源图像的SDD测度关于质子偏离场的Gateaux导定义为模型外力,将图像配准问题转换成力学平衡问题.针对模型演化过程中出现的网格重叠问题(即不拓扑保形),提出通过实时修改质子运动速度,而不是添加力的方法,保证模型演变具备拓扑保形性;为提高模型收敛速度,进一步将分层策略应用于模型演化过程中.若干的实验结果说明作者的工作是有意义的.
Image registration plays an important role in medical image processing, in this paper, a topology preservation image registration algorithm based on modified Spring-Mass model is presented. The image registration problem is transferred to the problem of mechanics balance, by defining the external force as the Gateaux derivate on mass displace field of template and source image＇ SSD measure. A scheme modifying the mass＇s velocity directly is used for the topology preservation of spring-mass modelrs evolving. Also, Multi-layers scheme is used for more efficient evolving. A series of experimental results are offered to show the useful of the method.
出处
《计算机学报》
EI
CSCD
北大核心
2008年第7期1224-1233,共10页
Chinese Journal of Computers
基金
国家杰出青年基金(60525213)
国家自然科学基金重点项目(60533030)
国家“九七三”重点基础研究发展规划项目基金(2006CB303106)
City U Applied R&D Funding(ARD)(9668009)
广东省自然科学基金(032202)
广东省科学基金(04300461)
广东省科技攻关项目(2004B33101005)资助
关键词
图像配准
弹簧质子模型
拓扑保形
image registration
spring-mass model
topology preservation
