Remote Sensing Image Data Fusion based on IHS and Local Deviation of Wavelet Transformation

Abstract

The goal of image fusion is to create new images that are more suitable for the purposes of human visual perception, machine vision, object detection and target recognition. The use of remote sensing image data such as SPOT Panchromatic and three-band Landsat multi-spectral images has led to increased recognition rate in applications such as automatic target recognition. In order to adequately make use of all kinds of remote sensing images information, a new image fusion method that combines IHS transform and local deviation of wavelet transform is proposed. Firstly, the multi-spectral image is transformed into the IHS components. Secondly, the histogram-matched panchromatic image and intensity component (I) are decomposed into wavelet coefficients respectively. Thirdly, a new intensity can be obtained by fusing the wavelet coefficient data of histogram-matched panchromatic image and intensity component (I) based on the local deviation of wavelet transform fusion method. Finally, the new intensity, together with the hue, saturation components, is transformed back into RGB space. The information entropy, the image clarity and the correlation coefficient are computed to assess this proposed method and some other methods. The simulation results illustrate that this new algorithm is more effective and the fused images are more suitable for human visual perception or machine vision