基于Gabor小波变换的无参考图像模糊度评价

Evaluation on No- reference Image Blur Metric Based on Gabor Wavelet Transform

当前的各种图像质量评价方法中,模糊是最常考虑的因素之一。针对模糊度量,基于Gabor小波提出了一种新的评价方法。首先利用Gabor小波提取图像高频信息,把原图像分割成高频和低频部分;然后分别对这两部分进行像素点误差累计,算出两部分的差值平均;最后结合这两个值算出模糊度。该方法结合了人类视觉系统模型和误差累计的思想。实验结果表明,模糊质量评价结果与人主观评价结果的相关系数为0.7836,异常值比为0.0443,说明这是一种有效的模糊评价方法。

The ambiguity is one of the factors most frequently considered in a variety of image quality assessment methods. This paper proposes a new evaluation method to detect the blur based on the Gabor wavelet. Firstly, the Gabor Wavelet is used to extract the high - frequency information of image. Then the original image is divided into high frequency and low frequency. Secondly, the two frequencies of image pixel point error are accumulated separately, and the average difference is calculated at the same time. Finally, the blur degree is calculated combining with these two values. The method combines the idea of the Human Visual System Model and cumulative error. The experimental results show that the correlation coefficient of the blur evaluation results and the manual evaluation results is 0. 7836 and the outlier ration is 0.0443, and the method is effective.