计算机工程 ›› 2012, Vol. 38 ›› Issue (3): 215-217.doi: 10.3969/j.issn.1000-3428.2012.03.072

• 图形图像处理 • 上一篇    下一篇

基于TV与SSIM的图像质量评价方法

庞璐璐a,李从利b,罗 军b   

  1. (解放军炮兵学院 a. 五系;b. 信息工程教研室,合肥 230031)
  • 收稿日期:2011-07-27 出版日期:2012-02-05 发布日期:2012-02-05
  • 作者简介:庞璐璐(1987-),女,硕士研究生,主研方向:图像处 理;李从利,副教授;罗 军,博士
  • 基金项目:
    国家自然科学基金资助项目(40876095)

Image Quality Assessment Method Based on TV and SSIM

PANG Lu-lu a, LI Cong-li b, LUO Jun b   

  1. (a. Five Department; b. Information Engineering T&R Section, Artillery Academy of PLA, Hefei 230031, China)
  • Received:2011-07-27 Online:2012-02-05 Published:2012-02-05

摘要: 提出一种基于全变分(TV)模型与结构相似度(SSIM)的图像质量评价方法。对待评价图像进行主动定量加噪,得到降质图像,利用自适应的TV去噪模型得到消噪图像,采用SSIM方法对待评价图像与消噪图像进行全参考评价,得到待评价图像的无参考评价指标。采用标准测试图像和LIVE库的降质图像进行实验,结果表明,该方法可在无参考图像的条件下对图像质量进行评估,评价结果与主观评价结果具有较高的一致性。

关键词: 图像质量评价, 全变分, 结构相似度, 人眼视觉系统, 图像去噪

Abstract: This paper proposes an image quality assessment method based on Total Variation(TV) model and Structural Similarity(SSIM). It adds noises to distorted image to quantitatively determine, and gets the degraded image. It uses the self-adaptive TV denoising model and gets denoising image, then uses the SSIM method to make reference evaluation between the distorted image and denoising image. The results is the no reference evaluating indicator. It uses the standard testing images and the degraded images from the LIVE database to make evaluate experiment, the results show that the method can judge the quality of images without explicit knowledge of the reference images, and it is highly consistent to the result of human visual.

Key words: image quality assessment, Total Variation(TV), Structural Similarity(SSIM), Human Visual System(HVS), image denoising

中图分类号: