| 基于改进光流场模型的大脑图像配准 |
| |
| 引用本文： | 白小晶,陈允杰,孙怀江,王平安,夏德深. 基于改进光流场模型的大脑图像配准[J]. 计算机辅助设计与图形学学报, 2008, 20(3): 349-355 |
| |
| 作者姓名： | 白小晶 陈允杰 孙怀江 王平安 夏德深 |
| |
| 作者单位： | 1. 南京理工大学计算机科学与技术学院,南京,210094
2. 香港中文大学计算机科学与工程学系,香港 |
| |
| 基金项目： | 香港研究资助局资助项目 , 香港中文大学校科研和教改项目 |
| |
| 摘 要： | 将光流场模型引入大脑图像配准,针对Horn模型会造成图像严重模糊的问题,在微分光流场模型的一般框架下,构造具有边缘保持和一致性增强能力的流驱动各向异性扩散方程作为正则项,以增强配准过程的特征保持能力;采用非二次惩罚函数作为数据项,以增强模型的鲁棒性.最后利用文中模型对大脑图像进行配准实验,得到了较为准确的结果.
|
| 关 键 词： | 大脑图像配准 流驱动各向异性正则化 非二次惩罚函数 边缘保持 一致性增强 |
| 收稿时间： | 2007-05-27 |
| 修稿时间： | 2007-12-19 |
An Improved Optical Flow Model for Brain Image Registration |
| |
| Bai Xiaojing,Chen Yunjie,Sun Huaijiang,Pheng Ann Heng,Xia Deshen. An Improved Optical Flow Model for Brain Image Registration[J]. Journal of Computer-Aided Design & Computer Graphics, 2008, 20(3): 349-355 |
| |
| Authors: | Bai Xiaojing Chen Yunjie Sun Huaijiang Pheng Ann Heng Xia Deshen |
| |
| Abstract: | An improved optical flow model within the differential framework is employed in brain image registration.Aiming at eliminating the severe image blurring caused by the Horn model,the anisotropic flow-driven diffusion,which is able to preserve edges and enhance coherence,is used as the regularization term to keep the image feature during the registration process.The data term employs the nonquadratic penalization function to improve the model's robustness.Our improved model is applied to register brain images,and more accurate and robust results are obtained. |
| |
| Keywords: | brain image registration anisotropic flow-driven regularization nonquadratic penalization edge preserving coherence enhancing |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
