基于小波变换的CT/PET图像融合最佳参数研究  

Research of Optimal Parameters of CT/PET Image Fusion Based on Wavelet Transform

作　　者：刘凌湘[1] 陈武凡[1] 周凌宏[1] 徐子海[2] 陈超敏[1] 

机构地区：[1]南方医科大学生物医学工程学院,广州510515 [2]解放军303医院放射治疗中心,南宁530021

出　　处：《中国生物医学工程学报》2010年第4期498-503,共6页Chinese Journal of Biomedical Engineering

基　　金：国家重点基础研究发展(973)计划(2010CB732500);广东省科技计划项目(2007B01040056)

摘　　要：为了提高基于小波变换图像融合的性能,在图像融合规则相对固定的情况下,提出一种确定最佳小波基函数和分解层数的方法。从图像的信息熵出发,通过比较低频子带图像熵差与原始图像熵差的接近程度,选择每一种小波基所对应的最佳分解层数;在小波分解层数确定的情况下,结合图像融合评价方法,选择最佳的小波基函数。与引入融合效果的评价构成一个闭环系统来确定小波参数相比,该方法极大地简化了判别过程;将该方法应用于CT/PET图像融合,获得了较好的融合效果。实验结果表明,该方法简单可行,对基于小波变换图像融合的小波参数选取有一定的指导意义。In order to improve the performance of image fusion based on wavelet transform,a new approach to determine the best wavelet base and decomposition level was proposed in the case of image fusion rule was ralatively fixed.First,according to the image information entropy,the best decomposition level of each wavelet base was selected by comparing the difference of both low-frequency sub-band image entropy and the original image entropy.After the best decomposition level was determined,the best wavelet base was chosen by image fusion evaluation methods.Compared with the method in which the parameters of wavelet decomposition were determined by introducing image fusion evaluation methods to form a closed-loop system,the proposed method greatly simplified the distinction process.The results of CT / PET image fusion were satisfying as well.The method is simple and feasible in selecting the wavelet parameters for wavelet transform based image fusion.

关 键 词：CT/PET 小波变换 图像融合 小波基 分解层数 

分 类 号：R814[医药卫生—影像医学与核医学]