Combination of Feature and Pixel Level Image Fusion with Feedback Retina and IHS Model

Abstract— The combined image fusion method which proposed in this paper integrates the information from multiple fusion algorithms which is more suitable for human visual perception. Multispectral images (SPOT) have low spatial resolution. So, integrating of low-resolution multispectral and high-resolution panchromatic (IKONOS) images is a technique for improving the spatial quality of multispectral images. An ideal fusion process preserves the original spectral characteristics and adds spatial characteristics to the image. The intensity-hue-saturation (IHS) algorithm and the feedback retina model fusion technique can maintain more spatial feature and more spectral information content, respectively. The presented algorithm integrates the advantages of both fusion methods to enhance the information content. Visual and statistical analyses show that the proposed algorithm significantly improves the fusion quality in terms of discrepancy, and average gradient; compared to fusion methods including, IHS, Brovey, discrete wavelet transform, atrous wavelet and Non-feedback retina model.