An adaptive fusion method based on regional feature for ALOS image

Abstract

Most Earth observation satellites provide both panchromatic images with a higher spatial resolution and multispectral images with a lower spatial resolution. Image fusion techniques can integrate the spatial detail of the panchromatic image and the spectral characteristics of the multispectral image into one image Exiting image fusion techniques such as the Intensity-Hue-Saturation (IHS) transform method, Brovey transform method and High Pass Filter (HPF) method may not be optimization while fusing the new generation commercial satellite image such as ALOS. The most serious problem is that the fused image usually has a notable deviation in visual appearance and spectral values from the original image. In this paper, we proposed a new fusion method for ALOS images, when adding the detail information of panchromatic image to the intensity component of multispectral image, the weight coefficients are determined adaptively based on the structural similarity(SSIM) between the panchromatic image and the intensity component of the multispectral image. Experimental results indicate that this method is effective when fusing ALOS image.