摘要
针对多光谱图像和全色图像的融合,提出一种基于显著区域分割和小波变换的遥感图像融合算法。首先通过IHS变换对多光谱图像进行分解,将亮度分量I和全色图像进行小波变换,得到对应的高低频系数。对低频系数进行模糊C均值聚类分析,并根据显著性因子分割出显著区域和非显著区域,针对不同区域和频段采用不同的准则进行融合。最后重建融合系数并进行IHS逆变换以获得融合图像。通过对比实验,表明该方法获得的融合结果优于传统的融合方法。
Aiming at the fusion of multi-spectral image and panchromatic image,a remote sensing image fusion algorithm based on salient region segmentation and wavelet transform is proposed. Firstly,the multi-spectral image is decomposed by IHS transform, and the intensity component I and the panchromatic image are decomposed by wavelet transform to obtain corresponding high and low frequency coefficients. Then the fuzzy C-means clustering analysis is performed on the low-frequency coefficients,and the salient regions and non-significant regions are segmented according to the significance factor,next different rule are used for the fusion of different regions and frequency bands. Finally,the fusion image is obtained by reconstructing the fusion coefficient and performing the IHS inverse transformation.The comparison experiments show that the fusion results are better than the traditional fusion methods.
作者
王帅
谢明鸿
黄秋萍
杨进
WANG Shuai;XIE Ming-hong;HUANG Qiu-ping;YANG Jin(Faculty of Information Engineering and Automation,Kunming University of Science and Technology,Kunming 650500,China)
出处
《信息技术》
2019年第1期57-60,共4页
Information Technology
基金
云南省科技计划项目(2017FB094)
关键词
图像融合
模糊C均值聚类
小波变换
显著性因子
image fusion
fuzzy C-means clustering
wavelet transform
significant factor
