全参考的图像质量评价总结

A Summary of the Full Reference Image Quality Evaluation

数字图像在实际的应用中会引入各种各样的失真,识别接受图像的是人的眼,所以图像质量的评价结果必须要与人眼的视觉效果相吻合。随着对人眼视觉系统的研究的不断深入与提高,研究人员根据HVS的特性建立图像质量评价模型来,这种模型大大的提高了图像质量主观评价和客观评价的一致性。文章详细地介绍了PSNR、SSIM、MSSIM、IFC、VIF、FSIM等几种典型的全参考图像质量评价方法分析探讨图像质量评价研究的发展趋势。

Digital images in the actual application will introduce a variety of distortion, recognition of the image of the eye, so the image quality evaluation results must be consistent with the visual effect of the human eye.With the study of human visual system（HVS） continues to deepen and improve, the researchers according to the HVS characteristics to establish the image quality evaluation model, this model greatly improves the consistency of subjective evaluation and objective evaluation of image quality. In this paper, the SSIM, MSSIM, PSNR, IFC, VIF, FSIM and other typical reference image quality evaluation methods are introduced, and the development trend of image quality evaluation is discussed.