基于改进小波变换的多聚焦医学图像融合方法  

Method of Multi-focal Medical Image Fusion Based on Improved Wavelet Transform

作　　者：许蓉 王直[1] 宗涛 陆蓉 杨莎莎 XU Rong;WANG Zhi;ZONG Tao;LU Rong;YANG Shasha(School of Computer Science,Jiangsu University of Science and Technology,Zhenjiang 212003;School of Electronic Information,Jiangsu University of Science and Technology,Zhenjiang 212003)

机构地区：[1]江苏科技大学计算机学院,镇江212003 [2]江苏科技大学电子信息学院,镇江212003

出　　处：《计算机与数字工程》2021年第8期1571-1573,1630,共4页Computer & Digital Engineering

摘　　要：为了提高多聚焦医学图像的融合质量,论文提出了一种基于小波域的改进融合算法。源图像经过小波变换分解为高频部分和低频部分,高频部分采用均方差取大的融合规则,低频部分采用基于prewitt算子和清晰度评价函数的融合规则,最后将新的高频和低频融合系数经过小波逆变换得到融合图像。通过与多种常用算法的融合效果图比较以及融合质量指标的分析,得出改进算法提高了融合后医学图像边缘和内部细节的清晰度。In order to improve the fusion quality of multi-focus medical images,this artical proposes an improved fusion algorithm based on wavelet domain.The source images are decomposed into high-frequency parts and low-frequency parts through the wavelet transformation.The high frequency part adopts a fusion rule which choose the maximal mean square deviation,and the low frequency part adopts the fusion rule based on prewitt operator and clarity evaluation function.Finally,the new fusion coefficients of high frequency and low frequency are obtained by inverse wavelet transform.Through the comparison with the fusion effect diagrams of many common algorithms and the analysis of the fusion quality index,it is concluded that the improved algorithm improves the clarity of the edge and internal details of the medical image after fusion.

关 键 词：均方差 PREWITT算子 清晰度评价函数 融合质量指标 

分 类 号：TN911.73[电子电信—通信与信息系统]