Share Email Print

Proceedings Paper

Color image quality assessment with biologically inspired feature and machine learning
Author(s): Cheng Deng; Dacheng Tao
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

In this paper, we present a new no-reference quality assessment metric for color images by using biologically inspired features (BIFs) and machine learning. In this metric, we first adopt a biologically inspired model to mimic the visual cortex and represent a color image based on BIFs which unifies color units, intensity units and C1 units. Then, in order to reduce the complexity and benefit the classification, the high dimensional features are projected to a low dimensional representation with manifold learning. Finally, a multiclass classification process is performed on this new low dimensional representation of the image and the quality assessment is based on the learned classification result in order to respect the one of the human observers. Instead of computing a final note, our method classifies the quality according to the quality scale recommended by the ITU. The preliminary results show that the developed metric can achieve good quality evaluation performance.

Paper Details

Date Published: 4 August 2010
PDF: 7 pages
Proc. SPIE 7744, Visual Communications and Image Processing 2010, 77440Y (4 August 2010); doi: 10.1117/12.863497
Show Author Affiliations
Cheng Deng, Xidian Univ. (China)
Dacheng Tao, Nanyang Technological Univ. (Singapore)

Published in SPIE Proceedings Vol. 7744:
Visual Communications and Image Processing 2010
Pascal Frossard; Houqiang Li; Feng Wu; Bernd Girod; Shipeng Li; Guo Wei, Editor(s)

© SPIE. Terms of Use
Back to Top