Abstract: Image Quality Assessment（IQA） has long been a hot topic in the research field of computer vision. It is still a hard-tackling problem to enable the computer to assess the image quality as human visual system. As a vital component of IQA, Image Definition Evaluation Function（IDEF） is not only of great significance, but also conductive to the development of the related fields such as image restoration and image enhancement. The current IDEFs are mainly based on the amount of the image edge information or overall information entropy to estimate the image’s definition. This paper discusses the general evaluation efficiency of five typical IDEFs, and aims to study on their effectiveness to evaluate the definition of different images blurred in different depth simultaneously, which will indicate the distance between the given five IDEFs and the subjective perception.

Key words: image quality assessment, image definition evaluation function, image edge, information entropy, general evaluation efficiency

摘要： 图像质量客观评价是计算机视觉领域长期以来研究的难点和热点问题。如何能让计算机像人的视觉系统一样具有评价图像质量的能力，至今仍是尚未完全解决的难题。图像清晰度评价函数是图像质量客观评价标准中的重要组成部分。现有的图像清晰度评价函数主要基于图像边缘细节或者整体信息熵的统计，给出对图像相对清晰度的估计。主要考察了五种具有代表性的清晰度评价函数，着重分析了它们对不同模糊程度下不同内容图像的清晰度评价能力，进而说明现有的清晰度客观评价方法与人主观感受的差距。

关键词: 图像质量客观评价, 清晰度评价函数, 图像边缘, 信息熵, 通用评价能力