首页 | 本学科首页   官方微博 | 高级检索  
     检索      

一种有效的形态小波多聚焦图像融合算法
引用本文:曹义亲,雷章明,黄晓生.一种有效的形态小波多聚焦图像融合算法[J].实验技术与管理,2012,29(3):38-41.
作者姓名:曹义亲  雷章明  黄晓生
作者单位:1. 华东交通大学软件学院,江西南昌,330013
2. 华东交通大学信息工程学院,江西南昌,330013
基金项目:江西省自然科学基金资助项目(2010GZS0025);江西省科技支撑计划项目(2008J212);江西省研究生创新专项资金项目(YZ2011-S082)
摘    要:多聚焦图像融合的关键问题是如何更好地保持源图像的轮廓信息和细节信息。形态小波具有小波变换的多层次分解特性和形态学的非线性特性。提出了一种基于形态小波变换的多聚焦图像融合算法。根据形态小波分解信号的特点,分别对高频系数和低频系数设计不同的融合规则:高频系数采用方向相关方法,最大程度保留细节和细节信息的方向特性;低频系数采用加权平均方法,该方法轮廓明显,最大程度地保留了轮廓信息。实验结果表明:该方法融合图像效果优于传统图像融合方法。

关 键 词:图像融合  形态小波  小波变换  多聚焦图像  数学形态学

An effective multi-focus image fusion algorithm based on morphological wavelets
Cao Yiqin , Lei Zhangming , Huang Xiaosheng.An effective multi-focus image fusion algorithm based on morphological wavelets[J].Experimental Technology and Management,2012,29(3):38-41.
Authors:Cao Yiqin  Lei Zhangming  Huang Xiaosheng
Institution:1.School of Software,East China Jiaotong University,Nanchang 330013,China; 2.School of Information Engineering,East China Jiaotong University,Nanchang 330013,China)
Abstract:The key point of multi-focus image fusion is to keep the outline and detail information of the original image better.As morphological wavelet integrates the properties of multi-level decomposition owned by wavelet transform and the non-linear features possessed by morphology,this paper proposes a multi-focus image fusion algorithm based on morphological wavelet.According to the characteristics of wavelet signal decomposition,this paper designs different fusion rules for high-frequency coefficients and low-frequency coefficients respectively.High-frequency coefficients use the method of achieving direction-related to reserve the details and orientation.However,the low-frequency coefficients adopt the method of weighted average to remain the obvious outline information maximally.The results indicate that the proposed image fusion method has better effects than traditional image fusion method.
Keywords:image fusion  morphological wavelet  wavelet transform  multi-focus image  mathematical morphology
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号