«上一篇/Previous Article|本期目录/Table of Contents|下一篇/Next Article»

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:

24

期数:

2014年03期

页码:

67-70

栏目:

智能、算法、系统工程

出版日期:

2014-03-31

文章信息/Info

Title:

An Improved Image Quality Assessment Method Based on Structural Similarity

文章编号:

1673-629X（2014）03-0067-04

作者:

糜靖峰¹; 李庆武¹²; 朱浩¹; 霍冠英¹²

1.河海大学 物联网工程学院；2.海洋与近海工程研究院

Author(s):

MI Jing-feng[1]; LI Qing-wu[1][2]; ZHU Hao[1]; HUO Guan-ying[1][2]

关键词:

结构相似度; 失真图像; 直方图集中度; 直方图结构相似度; 人眼视觉系统

Keywords:

SSIM; distorted image; histogram concentration; HSSIM; human visual system

分类号:

TP301

文献标志码:

A

摘要:

基于结构相似度(SSIM)的图像质量评价方法简单高效,准确性较高,评价性能优于峰值信噪比(PNSR)和均方误差( MSE),但SSIM模型不能较好地评价严重失真和交叉失真类型的图像。文中提出了一种改进的基于结构相似度的图像质量评价方法( HSSIM),该方法将直方图信息作为图像的主要结构信息,根据人眼视觉特性,利用直方图集中度来表示图像模糊度,最终计算得到图像的结构相似度值。实验结果表明,HSSIM比SSIM模型更符合人眼视觉系统特性,能更好地评价失真图像的质量。

Abstract:

SSIM is an image quality assessment algorithm with the advantage of simplicity,high efficiency and better consistence. Its eval-uation of performance is better than PNSR and MSE,however,it often fails when assessing badly distorted or cross distorted images. In this paper,an improved image quality assessment algorithm based on structural similarity ( HSSIM) is proposed,which takes the histo-gram concentration as the main structural information of an image,according to the human visual characteristics,using histogram concen-tration to calculate the fuzzy degree of the image,obtaining the structure similarity value of the image finally. Experimental results show that,compared with the SSIM model,the proposed HSSIM model is more consistent with human visual system and can assess the quality of fault images more precisely.

常用功能

更新日期/Last Update: 1900-01-01