基于多分辨率的多模态医学图像配准

J4 ›› 2012, Vol. 29 ›› Issue (1): 15-20.

基于多分辨率的多模态医学图像配准

马宾¹, 马德贵¹, 廖建敏²

1. 安徽农业大学, 安徽合肥, 230036
2. 广西机械工业研究院, 广西南宁, 530007

收稿日期:2010-11-26 修回日期:2011-07-08 出版日期:2012-01-28 发布日期:2012-01-28
通讯作者: 马宾（1977~），安徽庐江人，助教，研究方向为电子技术应用、图像采集与处理。 E-mail:mabin@ahau.edu.cn

Multimodality images registration based on multi-resolution

MA Bin¹, MA Degui¹, LIAO Jiangmin²

1.Anhui Agricultural University, Hefei 230036，China;
2. Guangxi Research Institute of Mechanical Industry, Nanning 530007，China

Received:2010-11-26 Revised:2011-07-08 Published:2012-01-28 Online:2012-01-28

摘要/Abstract

摘要：

提出了一种多分辨率多模态医学图像配准方法来提高图像的配准速度与精度。利用小波函数对图像进行多层分解，采用改进的Powell优化算法寻找待配准图像间的最大互信(MMI)，配准首先从图像金字塔中分辨率最低的0层图像开始，将0层图像配准获得的变形参数作为1层图像配准的起始参数，重复该过程直到完成最高精度层的图像配准。CT/MRI和CT/PET图像配准结果证明基于小波分解的多分辨率配准方法具有避免优化算子陷入局部极值、算子收敛速度快和配准精度高等优点。

关键词: 图像处理, 图像配准, 最大互信息, 小波, 改进Powell优化算法

Abstract:

The paper proposed a multi-resolution multi-modal medical image registration method to improve the speed and accuracy of registration. The images were decomposed by Wavelet function, and expanded Powell optimization algorithm was used to find the Maximize mutual information (MMI) between images. The registration starts from the lowest-resolution image level 0 of the images pyramid, the deformation parameters of image Level 0 registration will be available as the initial parameters of image Level 1, repeating the process until the completion of the highest- resolution image registration. The images registration results of CT/MRI and CT/PET show that wavelet-based multi-modal registration method has many advantages such as avoiding Optimizer getting into local maxima, fast convergence and high accuracy.

Key words: image processing, images registration, maximize mutual information, wavelets, expanded　Powell algorithm

中图分类号:

TP202.7

马宾, 马德贵, 廖建敏. 基于多分辨率的多模态医学图像配准[J]. J4, 2012, 29(1): 15-20.

MA Bin, MA Degui, LIAO Jiangmin. Multimodality images registration based on multi-resolution[J]. J4, 2012, 29(1): 15-20.

参考文献

[1] Zitova B, Flusser J. Image registration method :a survey[J]. Image and Vision Computing, 2003, 21(3):977-1000.
[2]Li Shixing, Guan Liang, Zhang Tiange, et al. Rigid Registration and Fusion between SPECT and MRI/CT Based on External Fiducial Alignment [J]. Chinese Journal of Medical Instrumentation(中国医疗器械杂志), 2001,25(4):192-195(in Chinese).
[3]Chen Ming, Chen Wufan. Contour based Rigid Registration of Multimodal Medical Images [J]. Journal of Image and Graphics(中国图象图形学报), 1999,3(10):854-857(in Chinese).
[4]Maes F, Vandermeulen D, Suetens P. Medical image registration using mutual information[C]. Proc. of the IEEE, 2003, 91(10)：1699- 1723.
[5]Luo Shuqian, Li Xiang. Multi-modality medical image registration based on maximization of mutual information [J]. Journal of Image and Graphics(中国图象图形学报), 2000, 5(7):551-558(in Chinese).
[6] Liu Zhexing, Li Shuxiang. Automatic serial image registration by regional counting in joint-histogram [J]. Journal of Image and Graphics(中国图象图形学报),2002,7(5):450-456(in Chinese).
[7]Peng Jinglin, Zhang Jing, Li Shutao, et al. Medical image registration based on expanded PV interpolation and hybrid optimization algorithm [J]. Acta Electronica Sinica(电子学报),2006,34(5):962-965(in Chinese).
[8]Fu Yili, Yu Xiaolong, Wang Yuehua. Fast algorithm of brain multi-modality image registration based on maximal mutual information [J]. Journal of Biomedical Engineering Research(生物医学工程研究), 2006, 25(2):71-74(in Chinese).
[9] Thevenaz P, Unser M.Optimization of mutual information for multiresolution image registration[J]. IEEE Trans. Image Processing, 2000, 9(12): 2083–2099.
[10] Broughton S A. Wavelet based methods in image processing[J]. Rose-Hulman Institute of Technology, Winter Quarter,1998.
[11] Powell M J D. An efficient method for finding the minimum of a function of several variables without calculating derivatives[J]. Comput. J,1964, 7:155-163.
[12] Tsao J. Interpolation artifacts in multimodality image registration based on maximization of mutual information[J]. IEEE Transactions on Medical Imaging, 2003, 22(7):854-864.

[1]	於康杰, 方波, 李剑敏, 王震, 蔡晋辉, 邬佳璐, 何正龙 . 太赫兹成像三维块匹配自适应Canny边缘检测算法[J]. 量子电子学报, 2023, 40(4): 458-468.
[2]	曾子威 , 李宏光 , 郭宇烽 , 廖文焘 . 隐匿危险品高准确度太赫兹光谱识别方法[J]. 量子电子学报, 2023, 40(3): 340-348.
[3]	葛宏义 , 王飞 , 蒋玉英 , 李丽 , 张元 , 贾柯柯 , . 基于宽度学习的太赫兹光谱图像小麦霉变识别研究[J]. 量子电子学报, 2023, 40(3): 360-368.
[4]	杜天宇, 高昆∗ , 吴朝. 光栅相衬 CT 中吸收信号环形伪影的去除方法[J]. 量子电子学报, 2023, 40(1): 40-47.
[5]	陈建明, , 曾祥津, , 钟丽云, , 邸江磊, ∗ , 秦玉文, ∗. 基于深度学习的图像配准方法研究进展综述[J]. 量子电子学报, 2022, 39(6): 899-926.
[6]	吴琼, 马雷∗. 一种基于 LoG 算子的量子图像边缘检测算法[J]. 量子电子学报, 2022, 39(5): 720-727.
[7]	杨一杉, 梁华为, 姚瑶, 陈帅∗. 金属遮挡条件下光子层析快速重构方法研究[J]. 量子电子学报, 2022, 39(4): 558-565.
[8]	张杨∗, 程正东, 朱斌. 基于改进 Faster R-CNN 的无人机小目标检测算法[J]. 量子电子学报, 2022, 39(3): 354-363.
[9]	毛仁祥, 常建华, 张树益李红旭, 张露瑶. 基于3D-MobileNetv2 的多目标实时跟踪框架[J]. 量子电子学报, 2022, 39(3): 364-372.
[10]	臧一鸣, 朱尚超, 魏战红∗, 刘志飞, 林小竹, 孙文韬. 一种量子图像伪彩色编码方法[J]. 量子电子学报, 2022, 39(3): 343-353.
[11]	章志涛, 丁芳, 罗宇, 陈夏华, 叶大为, 胡振华, 罗广南. 小波阈值去噪在偏滤器光谱信号处理中的应用[J]. 量子电子学报, 2022, 39(3): 307-315.
[12]	洪富祥, 陈冲, 丘仲锋∗. 一种基于机器视觉的无人机同心圆靶精准降落方法[J]. 量子电子学报, 2021, 38(3): 307-315.
[13]	陈儒, 杨义勤, 秦凡凯, 孟昭晖, 苗昕扬, 赵昆, 詹洪磊∗. 样本内部缺陷的太赫兹光谱成像与算法优化[J]. 量子电子学报, 2021, 38(3): 265-271.
[14]	王昱皓, 唐泽恬, 钟岷哲, 王阳, 曾瑞敏, 朱登玮, 杨晨. 基于相位相关和纹理分类的SIFT 图像拼接算法[J]. 量子电子学报, 2020, 37(6): 650-658.
[15]	王振武, 高隽, 范之国, 王昕, 张强. 利用正交偏振信息的透玻透膜成像方法[J]. 量子电子学报, 2019, 36(6): 691-698.