A multi-scale contrast-based image quality assessment model for multi-exposure image fusion

Abstract

In this paper, an accurate and efficient image quality assessment (IQA) model using the contrast information, called multi-scale contrast-based model (MCM), is proposed for conducting objective quality evaluation of multi-exposure image fusion (MEF). It is inspired by the fact that the human visual system (HVS) is highly sensitive to contrast that is naturally inherited in the MEF application. The key novelty of the proposed MCM lies in the usage of two salient contrast features, i.e., contrast structure and contrast saturation. For each reference and MEF images, the degree of similarity measured for each above-mentioned contrast attribute is then computed independently, followed by combining them together with the weight of each reference image computed based on its relevance to MEF image for obtaining contrast similarity maps (CSMs). Subsequently, all the obtained CSMs are fused using a standard deviation pooling strategy to generate the quality score. Finally, a multi-scale scheme is utilized to explore the image details from finer to coarser scales for producing the final MCM score. Simulation results have clearly shown that the proposed MCM model is more consistent with the perception of the HVS on the evaluation of MEF images than multiple state-of-the-art IQA methods.