| 全参考图像质量评价综述 |
| |
| 引用本文： | 褚 江,陈 强,杨曦晨. 全参考图像质量评价综述[J]. 计算机应用研究, 2014, 31(1): 13-22 |
| |
| 作者姓名： | 褚 江 陈 强 杨曦晨 |
| |
| 作者单位： | 南京理工大学 计算机科学与工程学院, 南京 210094 |
| |
| 基金项目： | 国家自然科学基金青年学者基金资助项目（60805003）; 南京理工大学自主科研专项基金资助项目（2011ZDJH26） |
| |
| 摘 要： | 图像质量评价是图像处理领域内一项很有意义的研究课题。客观图像质量评价方法可分为全参考评价方法、半参考评价方法和无参考评价方法, 目前全参考评价方法较为成熟, 而半参考和无参考评价方法则处于初级阶段, 远远达不到参考评价方法所能达到的效果。对全参考评价方法进行综述。首先简要地介绍了各种类型的评价方法, 其次详细地介绍了PSNR、SSIM、MSSIM、IFC、VIF、FSIM等几种典型的全参考图像质量评价方法, 然后在LIVE和TID2008数据库上进行实验, 对这几种全参考方法进行对比、分析, 最后探讨图像质量评价研究的发展趋势。
|
| 关 键 词： | 全参考图像质量评价 结构相似性 自然场景分析 特征相似性 |
Review on full reference image quality assessment algorithms |
| |
| CHU Jiang,CHEN Qiang,YANG Xi-chen. Review on full reference image quality assessment algorithms[J]. Application Research of Computers, 2014, 31(1): 13-22 |
| |
| Authors: | CHU Jiang CHEN Qiang YANG Xi-chen |
| |
| Affiliation: | School of Computer Science & Engineering, Nanjing University of Science & Technology, Nanjing 210094, China |
| |
| Abstract: | Image quality assessment is of fundamental importance to numerous image processing applications. Objective image quality metrics can be classified into three categories: full reference, semi-reference and no-reference. Currently, full reference metrics can achieve satisfied performance, while semi-reference and no-reference metrics are in their preliminary stage. This paper gave an overview of the full reference image quality assessment. Firstly, it introduced these methods briefly. Secondly, it described several important full reference image quality assessment algorithms in detail, such as PSNR, SSIM, MSSIM, IFC, VIF and FSIM. Then it compared the performance of these methods in the LIVE database and TID2008 database. Finally, this paper summarized the trends of future research on image quality assessment. |
| |
| Keywords: | full reference image quality assessment structure similarity natural scene statistics feature similarity |
|
| 点击此处可从《计算机应用研究》浏览原始摘要信息 |
|
点击此处可从《计算机应用研究》下载全文 |
|
