Abstract—Pan-sharpening has become a widely used tool to produce remotely sensed imagery which is rich in both spatial and spectral content. The main requirements on the pan-sharpening algorithms are to maximize the spatial content of the result and, simultaneously, to minimise spectral distortions, which may occur during processing. This paper presents a method for merging a set of multispectral bands with the higher spatially resolved panchromatic band. The experiments for the validation of the proposed pan-sharpening method were carried out with the use of data from modern very high resolution satellite instruments (IKONOS, THEOS, GeoEye-1). The results suggest that the spatial quality of the resulting images is significantly improved, while the spectral quality is maintained. Moreover, the actual merging algorithm is fast and straightforward to implement, since it mainly consists of linear operations with matrices.

Index Terms—Pan-sharpening, image fusion, satellite imagery, image processing.

Vladimir M. Buntilov is with the Department of Computer Engineering, Faculty of Engineering, Mahidol University, Phuttamonthon, Thailand, 73170 (e-mail: egbvd@mahidol.ac.th).

Cite:Vladimir M. Buntilov, "Pan-Sharpening of Multispectral Imagery Using Content-Specific Orthogonal Linear Transform," International Journal of Computer and Communication Engineering vol. 2, no. 3, pp. 260-265, 2013.