摘要
图像质量评价是计算机视觉和图像处理领域的一个热门问题,具有重要的研究价值和广阔的应用前景。根据所需原始图像信息的多少,客观图像质量评价方法主要分为全参考图像质量评价方法、部分参考图像质量评价方法和无参考图像质量评价方法。但是在实际应用过程中,参考图像往往难以获取或是不存在,因此需要对无参考图像质量评价算法进行研究。近年来,很多无参考质图像评算法被相继提出,但是国内还没有相关文献对这些算法的性能进行系统的分析和比较。基于此,论文首先选取了七种具有代表性的无参考图像质量评价算法并对其进行了简要介绍,然后全面介绍并总结了常用的图像质量评价算法性能验证数据库以及算法性能指标,最后,对七种算法的性能进行了详细的分析和比较。
Image quality assessment is a hot issue in the field of computer vision and image processing,and it has great research value and application prospect.According to the required knowledge of the original image,object image quality assessment methods are mainly divided into no-reference methods,reduced-reference methods and full-reference methods.But in practical application,the reference image is hard to achieve or even not exist.Therefore,it is necessary to do research in no-reference methods.Many no-reference methods have been proposed in recent years but no article has systematically analyze and compare these methods.Seven representative no-reference methods are chosen to do the job mentioned before and introduced some typical image quality databases and algorithm performance indicators.In the end,the seven algorithms are analyzed and compared in detail.
出处
《计算机与数字工程》
2016年第2期337-342,共6页
Computer & Digital Engineering
关键词
无参考
图像质量评价
图像质量数据库
算法性能
no-reference
image quality assessment
image quality database
algorithm performance