摘要
为了解决当前遥感图像融合算法存在的光谱失真以及模糊等问题,设计了基于Curvelet变换与双特征模型的遥感图像融合算法。利用IHS(intensity, hue, saturation)变换从多光谱图像中提取亮度分量。并将其与全色图像进行直方图匹配,以提高亮度分量与全色图像的相关性。随后,将亮度分量与全色图像进行Curvelet变换,以获取图像的低频、高频子带。利用低频子带的均值特征以及空间频率特征构造双特征模型,完成低频子带的融合。同时,利用高频子带的区域能量特征,完成高频子带的融合。将融合后的低频以及高频子带进行Curvelet以及IHS逆变换,输出融合图像。实验结果显示,与当前遥感图像融合算法相比,所提算法融合的遥感图像具有更好的光谱特性以及空间特性。
In order to solve the problems of spectral distortion and blurring existing in remote sensing image fusion algorithms,a remote sensing image fusion algorithm based on curvelet transform and dual feature model is designed and discussed in this paper. Firstly,the luminance component of the multispectral image was extracted by IHS transform. And the histogram was matched with panchromatic images to improve the correlation between luminance components and panchromatic images. Subsequently,the Curvelet transform was used to process the luminance component and panchromatic image to obtain the low frequency and high frequency sub-band of the image. The mean value feature and spatial frequency feature of low frequency subband were used to construct dual feature model for finishing the he fusion of low frequency subband. At the same time,the high frequency sub-band is fused by using the energy characteristics of the high frequency sub-band. Finally,the Curvelet and IHS inverse transform are used to get the final fused image. The experimental results show that this algorithm has better spectral and spatial characteristics than the current remote sensing image fusion algorithm.
作者
林宁
Lin Ning(College of Information Engineering,Nanning University,Nanning 530200,China)
出处
《电子测量与仪器学报》
CSCD
北大核心
2018年第12期78-84,共7页
Journal of Electronic Measurement and Instrumentation
基金
广西高校科学技术研究项目(KY2015YB525)
2018年度南宁学院科研项目(2018XJ33)资助
关键词
遥感图像融合
双特征模型
CURVELET变换
IHS变换
均值特征
空间频率特征
remote sensing image fusion
double feature model
curvelet transform
IHS transform
mean characteristic
spatial frequency characteristics
