TIQA-PSI: Toolbox for perceptual Image Quality Assessment of Pan-Sharpened Images

Abstract

A Pan-Sharpening (PS) technique integrates spatial details of a high-resolution panchromatic (PAN) image with spectral information of a low-resolution multi-spectral (MS) image, creating a single high-resolution color image. Since different PS methods produce images of a different quality, they should be compared using Image Quality Assessment (IQA) approaches that mimics human visual perception. In this paper, a MATLAB-based Toolbox for perceptual Image Quality Assessment of Pan-Sharpened Images (TIQA-PSI) is presented. TIQA-PSI aims to stimulate the development and facilitate the evaluation of new PS or PS-IQA approaches as it contains a large number of state-of-the-art IQA methods and images generated by representative PS techniques with corresponding subjective scores. They can be used to assess the output of any new PS algorithm or to serve as baselines for the comparison with new PS-IQA methods. Furthermore, apart from popular PS-IQA methods, the toolbox offers an opportunity to employ perceptual IQA approaches that can be effectively trained on scores obtained in tests with human subjects for superior image evaluation. The code of TIQA-PSI is available on GitHub.

Keywords

• Image Quality Assessment
• Hyperspectral Imaging
• Pan-sharpening
• Toolbox
• Perceptual image quality
• MATLAB