基于Contourlet和IHS的遥感图像融合

针对目前最新发展的Contourlet变换较小波变换能提供更丰富的方向和形状,有助于捕捉图像中的几何结构,提出了一种新的基于Contourlet变换和IHS(Intensity-Hue-Saturation)变换的遥感图像融合方法,首先对多频谱图像进行IHS变换,然后对所得的亮度分量和全色图像分别进行Contourlet变换,再对得到的低频近似系数和高频细节系数采用一定的融合规则得到一个新的亮度分量,并对其做逆向的IHS变换得到融合图像.实验结果表明,该方法在保留多频谱图像的频谱信息的同时增强了融合图像的空间细节表现能力,提高了融合图像的信息量,并且优于同等条件下的小波变换方法,该方法是有效可行的.

Remote Sensing Image Fusion Based on Contourlet and IHS

Remote sensing image fusion is a very important issue in remote sensing.Recently developed contourlet transform can offer a rich set of directions and shapes,thus it is more effective than wavelet in capturing smooth contours and geometric structures in images.This paper proposes a novel method of fusing panchromatic and multispectral images based on contourlet and IHS.The I(Intensity) component obtained from multispectral image through IHS transform and the panchromatic image are decomposed by contourlet transform to get low frequency coefficients and high frequency coefficients respectively,then fuse the coefficients according to a certain fusion rule to get a new I component,finally the fused image is obtained through inverse IHS transform.The experimental results show that the proposed method can effectively preserve spectral information and improve spatial resolution and information of the fused images,and outperforms the wavelet method at the same condition.