An image quality assessment algorithm based on feature selection

Ting Lu, Yanning Zhang, Haisen Li

科研成果: 书/报告/会议事项章节会议稿件同行评审

1 引用 (Scopus)

摘要

Image Quality Assessment(IQA) is of fundamental importance to numerous imaging and video processing applications. For most of the applications, the perceptual meaningful measure is the one which can automatically assess the quality of images or videos in a perceptually consistent manner. However, most commonly used IQA metrics are not consistent well with the human judgments of image quality. Recently, the SSIM metric which takes people's visual characteristics into consideration performs much better than the traditional PSNR/MSE. But the defects of it still exit on some specific kinds of distortions. A new algorithm of IQA based on feature selection is proposed in this paper. Local gradient entropy and phase congruency are added to the SSIM framework. Through in-depth feature selection and definition plus better pooling strategy, this algorithm performs much better in LIVE datasets.

源语言英语
主期刊名Intelligent Science and Intelligent Data Engineering - Third Sino-Foreign-Interchange Workshop, IScIDE 2012, Revised Selected Papers
289-297
页数9
DOI
出版状态已出版 - 2013
活动3rd Sino-Foreign-Interchange Workshop on Intelligent Science and Intelligent Data Engineering, IScIDE 2012 - Nanjing, 中国
期限: 15 10月 201217 10月 2012

出版系列

姓名Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
7751 LNCS
ISSN(印刷版)0302-9743
ISSN(电子版)1611-3349

会议

会议3rd Sino-Foreign-Interchange Workshop on Intelligent Science and Intelligent Data Engineering, IScIDE 2012
国家/地区中国
Nanjing
时期15/10/1217/10/12

指纹

探究 'An image quality assessment algorithm based on feature selection' 的科研主题。它们共同构成独一无二的指纹。

引用此