Circular-ELM for the reduced-reference assessment of perceived image quality

被引:54
作者
Decherchi, Sergio [1 ]
Gastaldo, Paolo
Zunino, Rodolfo
Cambria, Erik [3 ]
Redi, Judith [2 ]
机构
[1] Fdn Ist Italiano Tecnol IIT, Dept Drug Discovery & Dev, Genoa, Italy
[2] Delft Univ Technol, Dept Intelligent Syst, NL-2628 CD Delft, Netherlands
[3] Natl Univ Singapore, Temasek Labs, Singapore 117411, Singapore
基金
英国工程与自然科学研究理事会;
关键词
Extreme learning machine; Circular backpropagation; Image quality assessment; EXTREME LEARNING-MACHINE; CLASSIFICATION; NETWORKS;
D O I
10.1016/j.neucom.2011.12.050
中图分类号
TP18 [人工智能理论];
学科分类号
140502 [人工智能];
摘要
Providing a satisfactory visual experience is one of the main goals for present-day electronic multimedia devices. All the enabling technologies for storage, transmission, compression, rendering should preserve, and possibly enhance, the quality of the video signal; to do so, quality control mechanisms are required. These mechanisms rely on systems that can assess the visual quality of the incoming signal consistently with human perception. Computational Intelligence (CI) paradigms represent a suitable technology to tackle this challenging problem. The present research introduces an augmented version of the basic Extreme Learning Machine (ELM), the Circular-ELM (C-ELM), which proves effective in addressing the visual quality assessment problem. The C-ELM model derives from the original Circular BackPropagation (CBP) architecture, in which the input vector of a conventional MultiLayer Perceptron (MLP) is augmented by one additional dimension, the circular input; this paper shows that C-ELM can actually benefit from the enhancement provided by the circular input without losing any of the fruitful properties that characterize the basic ELM framework. In the proposed framework, C-ELM handles the actual mapping of visual signals into quality scores, successfully reproducing perceptual mechanisms. Its effectiveness is proved on recognized benchmarks and for four different types of distortions. (C) 2012 Elsevier B.V. All rights reserved.
引用
收藏
页码:78 / 89
页数:12
相关论文
共 40 条
[1]
[Anonymous], 1986, FOUNDATIONS, DOI DOI 10.7551/MITPRESS/5236.001.0001
[2]
[Anonymous], 1993, THESIS BROWN U PROVI
[3]
No-reference JPEG-image quality assessment using GAP-RBF [J].
Babu, R. Venkatesh ;
Suresh, S. ;
Perkis, Andrew .
SIGNAL PROCESSING, 2007, 87 (06) :1493-1503
[4]
Bartlett P., 2011, MACH LEARN, V48, P85
[5]
Sentic Album: Content-, Concept-, and Context-Based Online Personal Photo Management System [J].
Cambria, Erik ;
Hussain, Amir .
COGNITIVE COMPUTATION, 2012, 4 (04) :477-496
[6]
Objective quality assessment of color images based on a generic perceptual reduced reference [J].
Carnec, Mathieu ;
Le Callet, Patrick ;
Barba, Dominique .
SIGNAL PROCESSING-IMAGE COMMUNICATION, 2008, 23 (04) :239-256
[7]
Tactile-Data Classification of Contact Materials Using Computational Intelligence [J].
Decherchi, Sergio ;
Gastaldo, Paolo ;
Dahiya, Ravinder S. ;
Valle, Maurizio ;
Zunino, Rodolfo .
IEEE TRANSACTIONS ON ROBOTICS, 2011, 27 (03) :635-639
[8]
Image quality measures and their performance [J].
Eskicioglu, AM ;
Fisher, PS .
IEEE TRANSACTIONS ON COMMUNICATIONS, 1995, 43 (12) :2959-2965
[9]
Regularization networks and support vector machines [J].
Evgeniou, T ;
Pontil, M ;
Poggio, T .
ADVANCES IN COMPUTATIONAL MATHEMATICS, 2000, 13 (01) :1-50
[10]
Neural networks for the no-reference assessment of perceived quality [J].
Gastaldo, P ;
Zunino, R .
JOURNAL OF ELECTRONIC IMAGING, 2005, 14 (03) :1-11