Image Quality Assessment Based on Low Order Moment Features
Leida Li, Hancheng Zhu, Gaobo Yang · IEICE Transactions on Fundamentals of Electronics Communications and Computer Sciences · 2014
This letter presents a new image quality metric using low order discrete orthogonal moments. The moment features are extracted in a block manner and the relative moment differences (RMD) are computed. A new exponential function based on RMD is proposed to generate the quality score. The performance of the proposed method is evaluated on public databases. Experimental results and comparisons demonstrate the efficiency of the proposed method.