基于假设检验的图像质量的度量方法

An Image Quality Evaluation Method Based on Hypothesis Test

图像质量的正确评价是图像信息工程领域里一项很有意义的研究课题,但是现有的图像客观评价方法并不完全符合人的视觉特性.基于人的视觉特征和误差的结构与分布的特性,提出了对编码图像质量尺度假设检验的方法,结果显示,该方法能有效地度量图像的质量.

A correct evaluation of image quality is a very significant research subject in image information processing. But the existing objective evaluation methods for image quality are not in accordance with human visual characteristic. In this paper, a novel image quality evaluation method is presented, taking into account both the properties of visual perception and the features of error structure and distribution, and applying hypothesis test to coding image. The results show this method can evaluate image quality effectively.