A color image quality assessment using a reduced-reference image machine learning expert

Christophe Charrier (1) , Gilles Lebrun (2) , Olivier Lezoray (1)

1 Equipe Image - Laboratoire GREYC - UMR6072
2 GREYC - Groupe de Recherche en Informatique, Image et Instrumentation de Caen

A quality metric based on a classification process is introduced. The main idea of the proposed method is to avoid the error pooling step of many factors (in frequential and spatial domain) commonly applied to obtain a final quality score. A classification process based on the Support Vector Machine method is designed to obtain the final quality class with respect to the standard quality scale provided by the UIT. Thus, for each degraded color image, a feature vector is computed including several Human Visual System characteristics, such as, contrast masking effect, color correlation, and so on. Selected features are of two kinds: 1) full-reference features and 2) no-reference characteristics. That way, a machine learning expert, providing a final class number is designed.

Domaines: Traitement des images [eess.IV]

Christophe Charrier, Gilles Lebrun, Olivier Lezoray. A color image quality assessment using a reduced-reference image machine learning expert. SPIE Electronic Imaging, 2008, United States. ⟨10.1117/12.766473⟩. ⟨hal-00329517⟩