A Comparative Study of Quality Assessement Techniques for Pansharpening of High Resolution Satellite Imagery

A Comparative Study of Quality Assessement Techniques for Pansharpening of High Resolution Satellite Imagery

Soroosh Mehravar1 Ahmad Toosi2 Farhad Samadzadegan3 Farzaneh DadrasJavan4

1) Master student of Photogrammetry from Tehran University
2) Master student of Photogrammetry from Tehran University
3) Doctor of Photogrammetry from Tehran University
4) Doctor of Photogrammetry from Tehran University

IOI Code : XBAE-DZGEE
Publication : Demo - Website of Saman Conference Management System(ngtu.ir)
Abstract :
    In order to investigate the performance of the spectral and spatial quality assessment metrics in detecting the registration error of input satellite images of pan-sharpening process, some regions in the input images, the fused images and the spectral and spatial distorted images were selected, and after applying the metrics, the amount of change for each index have been calculated. Using this approach, the ability of each metric in detecting the registration error of input images was recognized. By perusing the obtained results, SAM and QNR were considered as the best and worst spectral indices, respectively, due to the degree of sensitivity to the registration error of input images, while Ds and HP-CC have shown the best and worst performance in displaying the spatial effect of the mentioned error, respectively.

Keywords : satellite imagery image fusion pansharpening spatial and spectral quality comparative study