Comparative analysis of pan-sharpening techniques on DubaiSat-1 images

This paper evaluates the performance of a set of pan-sharpening methods on DubaiSat-1 images. DubaiSat-1 is a new satellite and the evaluation of pan-sharpening methods shall promote new applications of data. Methods are selected to represent different approaches of pan-sharpening. The methods are the generalized Intensity-Hue-Saturation method, the principle component substitution method, and Gram-Schmidt method from the component substitution category, Brovey transform method, University of New Brunswick method, and smoothing filter based intensity modulation method from the modulation-based category, and basic high-pass filtering method, substitutive wavelet method, and additive wavelet luminance proportional method from the filtering-based category. The pan-sharpened images are quantitatively evaluated for their spatial and spectral quality using a set of well-established measures in the field of remote sensing. The evaluation metrics are ERGAS, Q4, and SAM which measure the spectral quality and a Laplacian-based metric that measures the spatial quality. Results show that images pan-sharpened using additive wavelet luminance proportional method are the best in terms of spatial and spectral quality.