An Image Quality Assessment Method Based on Granular Computing and Grey Relational Analysis

摘要

A Granular Computing (GrC) and Grey Relational Analysis (GRA) based method for image quality assessment is proposed in this paper. In this method, the original image is converted to some wavelet granules via Discrete Wavelet Packet Transform (DWPT). The conversion and combination ofthese granules are further discussed within the framework of quotient space theory. In addition, GRA of Grey theory is used to evaluate the image quality at different granularity levels, and an overall quality order is finally obtained by indexing the images on their grey granular quality.With the experimental results from LIVE image database, it is found that the proposed method performances best in terms of stability, suitability, and consistency with Human Visual System (HVS) and theoretical analysis.