No-reference blur image quality metric combining HVS with SSIM
Wanli Yuan, Li Chaofeng · Computer Engineering and Applications Journal · 2013
The traditional no-reference blur metrics haven’t considered the limitation of the Human Visual System(HVS)in blur detection. A novel no-reference blur metric combining characteristics of HVS with Structure Similarity(SSIM)is proposed. In the new method, a new re-blurred image is produced by convoluting the original image with a low pass filter. Then a collection of strong edge block around edge point detected by the Sobel operator is created. For each pixel block, the SSIM index of it and corresponding pixel block in re-blurred image is calculated. At last, the new blur metric as the average of the SSIM indices is taken. The experimental results on the LIVE blur database demonstrate that the method can obtain good performances. The linear correlation coefficient and the Spearman rank correlation coefficient between the results of the proposed method and the subjective quality measurements are 0.929 8 and 0.931 8 respectively.