Comparison of adaptive filter, IHS and Fourier transform based techniques for IRS-1C PAN/LISS-III data merging

This paper proposes two techniques for merging of IRS-1C LISS-III and panchromatic images based on Fourier transform and adaptive filtering. Performances of the proposed data merging techniques have been assessed quantitatively along with that of intensity-hue-saturation (IHS) domain data merging using Sobel edge intensity and normalized difference vegetation index (NDVI). Frequency domain merging is found to give minimum spatial distortion, while IHS merging yields the minimum spectral loss.