Image Quality Assessment with Manifold and Machine Learning - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2009

Image Quality Assessment with Manifold and Machine Learning

Résumé

A crucial step in image compression is the evaluation of its performance, and more precisely the available way to measure the fi nal quality of the compressed image. In this paper, a machine learning expert, providing a final class number is designed. The quality measure is based on a learned classi fication process in order to respect the one of human observers. Instead of computing a fi nal note, our method classi fies the quality using the quality scale recommended by the UIT. This quality scale contains 5 ranks ordered from 1 (the worst quality) to 5 (the best quality). This was done constructing a vector containing many visual attributes. Finally, the final features vector contains more than 40 attibutes. Unfortunatley, no study about the existing interactions between the used visual attributes has been done. A feature selection algorithm could be interesting but the selection is highly related to the further used classi fier. Therefore, we prefer to perform dimensionality reduction instead of feature selection. Manifold Learning methods are used to provide a low-dimensional new representation from the initial high dimensional feature space. The classi cation process is performed on this new low-dimensional representation of the images. Obtained results are compared to the one obtained without applying the dimension reduction process to judge the effi ciency of the method.
Fichier principal
Vignette du fichier
Charrier_EI09.pdf (198.59 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-00363265 , version 1 (21-01-2014)

Identifiants

Citer

Christophe Charrier, Gilles Lebrun, Olivier Lezoray. Image Quality Assessment with Manifold and Machine Learning. Image Quality and System Performance VI, Jan 2009, San Jose, United States. 11 pp, ⟨10.1117/12.810164⟩. ⟨hal-00363265⟩
161 Consultations
113 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More