Objective quality assessment of displayed images by using neural networks

被引:40
作者
Gastaldo, P
Zunino, R
Heynderickx, I
Vicario, E
机构
[1] Univ Genoa, DIBE, I-16145 Genoa, Italy
[2] Philips Res Labs, NL-5656 AA Eindhoven, Netherlands
关键词
perceptual quality; objective image quality; neural networks;
D O I
10.1016/j.image.2005.03.013
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Considerable research effort is being devoted to the development of image-enhancement algorithms, which improve the quality of displayed digital pictures. Reliable methods for measuring perceived image quality are needed to evaluate the performances of those algorithms, and such measurements require a univariant (i.e., no-reference) approach. The system presented in this paper applies concepts derived from computational intelligence, and supports an objective quality-assessment method based on a circular back-propagation (CBP) neural model. The network is trained to predict quality ratings, as scored by human assessors, from numerical features that characterize images. As such, the method aims at reproducing perceived image quality, rather than defining a comprehensive model of the human visual system. The connectionist approach allows one to decouple the task of feature selection from the consequent mapping of features into an objective quality score. Experimental results on the perceptual effects of a family of contrast-enhancement algorithms confirm the method effectiveness, as the system renders quite accurately the image quality perceived by human assessors. (c) 2005 Elsevier B.V. All rights reserved.
引用
收藏
页码:643 / 661
页数:19
相关论文
共 39 条
[31]   Empirical measure of multiclass generalization performance: The K-winner machine case [J].
Ridella, S ;
Zunino, R .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2001, 12 (06) :1525-1529
[32]   Circular backpropagation networks for classification [J].
Ridella, S ;
Rovetta, S ;
Zunino, R .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 1997, 8 (01) :84-97
[33]   Human vision and electronic imaging [J].
Rogowitz, BE ;
Pappas, TN ;
Allebach, JP .
JOURNAL OF ELECTRONIC IMAGING, 2001, 10 (01) :10-19
[34]   A multiprocessor-oriented visual tracking system [J].
Rovetta, S ;
Zunino, R .
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 1999, 46 (04) :842-850
[35]  
Rumelhart DE, 1986, PARALLEL DISTRIBUTED, V1, DOI DOI 10.7551/MITPRESS/5236.001.0001
[36]   No reference PSNR estimation for compressed pictures [J].
Turaga, DS ;
Chen, YW ;
Caviedes, J .
SIGNAL PROCESSING-IMAGE COMMUNICATION, 2004, 19 (02) :173-184
[37]   ACCELERATING THE CONVERGENCE OF THE BACK-PROPAGATION METHOD [J].
VOGL, TP ;
MANGIS, JK ;
RIGLER, AK ;
ZINK, WT ;
ALKON, DL .
BIOLOGICAL CYBERNETICS, 1988, 59 (4-5) :257-263
[38]   A universal image quality index [J].
Wang, Z ;
Bovik, AC .
IEEE SIGNAL PROCESSING LETTERS, 2002, 9 (03) :81-84
[39]   30 YEARS OF ADAPTIVE NEURAL NETWORKS - PERCEPTRON, MADALINE, AND BACKPROPAGATION [J].
WIDROW, B ;
LEHR, MA .
PROCEEDINGS OF THE IEEE, 1990, 78 (09) :1415-1442