一种改进的全参考模糊度比值算法

由于图像处理技术的发展, 对图像和视频评价提出了更高的要求, 而模糊度（blur）指标可以比较准确地反映图像的客观质量。为了提高图像视频客观评价的准确率, 针对模糊度进行了研究, 提出了一种改进的全参考模糊度比值算法。算法先使用边缘检测得到视频每一帧图像的垂直边缘图, 然后计算其模糊度值, 再将原始视频和失真视频模糊度值在坐标中标出, 一般情况下, 有点状、线状和散点分布三种, 最后使用三种方案：均值、线性回归和最小包容圆对其分布进行拟合。实验结果表明, 这三种方案所得的客观评价结果和主观评价值均有较好的吻合度。

Improved algorithm based on full reference blur ratio

Due to the development of the image processing technology, it made higher requests for image and video evaluation, and blur metric could reflect the objective quality of image and video more accurately. In order to improve the accuracy of objective evaluation of image and video, this paper studied the blur and proposed an improved algorithm based on full reference blur ratio. It first used edge detection algorithm to get the vertical edge image and calculate blur value of every frame of video, then marked the blur value in the coordinate of the original and distortion video, as usual, there were three kinds of distributions:dot, line and scatter, at last, used three solutions to match the distributions:slop of mean, linear regression and minimum enclosing circle, the result of experiment shows that these three kinds of solutions can give higher reference between subjective evaluation and objective evaluation of image and video.