A New Intensity-Hue-Saturation Based Method for Fusing High Resolution Satellite Images
要旨Among various image fusion techniques, the intensity-hue-saturation (IHS) based methods are widely used in many commercial software packages to merge high spatial resolution Panchromatic (PAN) images with high spectral resolution multispectral (MS) images. Several downsides are inherent with IHS-based fusion methods such as significant color distortions, low spatial quality and complicated computations especially when merging large volumes of Ikonos or QuickBird images. This paper introduces a new IHS-based fusion method to overcome these downsides. In the proposed method a spectrally adjusted intensity (I) image derived by introducing weighting and tradeoff parameters to all the MS bands was used. The weighting parameters were assigned according to the intersection area between the spectral response curve of each MS band and that of the PAN image. The tradeoff parameter was then used with different values to control the tradeoff between the spatial and spectral information of the images to be fused. The proposed method was applied to merge PAN and MS images of QuickBird and Ikonos covering different features of Tanta and Alexandria cities in Egypt. The fused images resulted by the new method were compared visually and statistically to the original PAN and MS images as well as to those resulted by other existing IHS-based fusion methods. In addition to its fast and simple implementation, the proposed method has proved its capability to provide more accurate fused images better than the other existing IHS-based fusion methods, both spatially and spectrally.