Title :
A new four-component gradient-based structural similarity metric using adaptive weights
Author :
Shanbo Gu ; Feng Shao ; Gangyi Jiang ; Mei Yu
Author_Institution :
Fac. of Inf. Sci. & Eng., Ningbo Univ., Ningbo, China
Abstract :
In this paper, a novel four-component weighted gradient-based structural similarity metric is proposed, according to the effects on the edges, smooth and texture for different types of distortion, by adaptively adjusting the weights of four components (strong edge, weak edge, texture region, and smooth region). Experiments show that the proposed metric can provide better assessment performance compared with other classical image quality assessment metrics.
Keywords :
gradient methods; image processing; adaptive weight; four-component structural similarity metric; gradient-based structural similarity metric; image distortion type; image quality assessment metrics; smooth region component; strong edge component; texture region component; weak edge component; Humans; Image edge detection; Image quality; Measurement; Transform coding; Visualization; White noise; adaptive weight; image quality assessment; structural similarity;
Conference_Titel :
Fuzzy Systems and Knowledge Discovery (FSKD), 2011 Eighth International Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-1-61284-180-9
DOI :
10.1109/FSKD.2011.6019662