28 April 2018 Predicting perceptual quality of images in realistic scenario using deep filter banks
Author Affiliations +
Abstract
Classical image perceptual quality assessment models usually resort to natural scene statistic methods, which are based on an assumption that certain reliable statistical regularities hold on undistorted images and will be corrupted by introduced distortions. However, these models usually fail to accurately predict degradation severity of images in realistic scenarios since complex, multiple, and interactive authentic distortions usually appear on them. We propose a quality prediction model based on convolutional neural network. Quality-aware features extracted from filter banks of multiple convolutional layers are aggregated into the image representation. Furthermore, an easy-to-implement and effective feature selection strategy is used to further refine the image representation and finally a linear support vector regression model is trained to map image representation into images’ subjective perceptual quality scores. The experimental results on benchmark databases present the effectiveness and generalizability of the proposed model.
© 2018 SPIE and IS&T
Weixia Zhang, Weixia Zhang, Jia Yan, Jia Yan, Shiyong Hu, Shiyong Hu, Yang Ma, Yang Ma, Dexiang Deng, Dexiang Deng, } "Predicting perceptual quality of images in realistic scenario using deep filter banks," Journal of Electronic Imaging 27(2), 023037 (28 April 2018). https://doi.org/10.1117/1.JEI.27.2.023037 . Submission: Received: 29 November 2017; Accepted: 5 April 2018
Received: 29 November 2017; Accepted: 5 April 2018; Published: 28 April 2018
JOURNAL ARTICLE
12 PAGES


SHARE
RELATED CONTENT

LoG acts as a good feature in the task of...
Proceedings of SPIE (March 06 2014)
An ICA-based approach for image quality assessment
Proceedings of SPIE (October 29 2009)
Personalized numerical observer
Proceedings of SPIE (February 26 2010)
Robust language-independent OCR system
Proceedings of SPIE (January 28 1999)

Back to Top