基于区域能量和层间相关性的图像加权融合算法  

Image Weighted Fusion Algorithm Based on Regional Energy and Layers Correlation

作　　者：李红岩[1] 宁梓淯[1] 李可[1] 李蒙[1] 

机构地区：[1]哈尔滨医科大学(大庆)医学信息学院,黑龙江大庆163319

出　　处：《哈尔滨理工大学学报》2014年第5期99-102,共4页Journal of Harbin University of Science and Technology

基　　金：黑龙江省卫生厅科学研究项目(2011-232)

摘　　要：研究了一种基于CL多小波变换的区域局部能量的多模态医学图像的融合方法.该方法改进了以往依靠能量匹配度的方法,充分研究了图像局部能量之间的数值关系.当多小波分解后,用图像层间相关系数来确定融合权值.该算法稳定性强,基本满足诊断的实时性要求.实验结果表明,该方法相较于单纯GHM区域能量融合方法,不仅将图像的信息熵和互信息熵,分别提高了2%和5%,交叉熵和标准差等指标也不同程度的有所提高,而且有较好的直观视觉效果.In this paper,we study a kind of image fusion methods,which based on local energy of CL multiwavelet transform.This method improved energy fusion rely on energy matching degree,and studied the numerical relation of the multi-layer local energy fully.Image correlation coefficient between the layers is used to determine the fusion weight after multiple wavelet decomposition.The algorithm stability is strong,easy to meet the real-time requirement the diagnosis.Experimental results show that the method is compared with pure GHM regional energy fusion method,not only the image information entropy and mutual information entropy,increased by 2％ and 5％ respectively,the cross entropy and standard deviation also improved.Have a good intuitive visual effects.

关 键 词：CL多小波 图象融合 区域能量 层间相关性 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]