FR-IQA - Permutation Entropy Deviation Index

In this work, a new objective image quality assessment (IQA) framework, based on the working principle of permutation entropy (PE) is proposed. The framework is titled as Permutation Entropy Deviation Index (PEDI). The idea is to design an IQA framework that should be highly accurate as well as computationally efficient; making it viable to be used with different image processing applications, for visual quality assessments. The proposed model make use of the PE that helps in detecting and visualizing changes related to structures with correlation between successive samples instead of considering magnitudes of the signal. Thus, the proposed approach uses this property to efficiently predict image quality. The PE exploits the global variations in the local quality map for image quality assessment. With standard deviation as the pooling strategy, it is noted that permutation entropy between reference and distorted images can predict image quality with high measures of accuracy. Experimental results on subjective database, CSIQ, have shown that the proposed model outperforms most of existing SOTA image quality assessment models and highly correlates with subjective judgements.