Blur is one of the most common distortion type and greatly impacts image quality. Most existing no-reference (NR) image blur metrics produce scores without a fixed range, so it is hard to judge the extent of blur directly. This letter presents a NR perceptual blur metric using Saliency Guided Gradient Similarity (SGGS), which produces blur scores with a fixed range of (0,1). A blurred image is first reblurred using a Gaussian low-pass filter, producing a heavily blurred image. With this reblurred image as reference, a local blur map is generated by computing the gradient similarity. Finally, visual saliency is employed in the pooling to adapt to the characteristics of the human visual system (HVS). The proposed metric features fixed range, fast computation and better consistency with the HVS. Experiments demonstrate its advantages.
Peipei ZHAO
China University of Mining and Technology
Leida LI
China University of Mining and Technology
Hao CAI
China University of Mining and Technology
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Peipei ZHAO, Leida LI, Hao CAI, "Saliency Guided Gradient Similarity for Fast Perceptual Blur Assessment" in IEICE TRANSACTIONS on Information,
vol. E98-D, no. 8, pp. 1613-1616, August 2015, doi: 10.1587/transinf.2015EDL8041.
Abstract: Blur is one of the most common distortion type and greatly impacts image quality. Most existing no-reference (NR) image blur metrics produce scores without a fixed range, so it is hard to judge the extent of blur directly. This letter presents a NR perceptual blur metric using Saliency Guided Gradient Similarity (SGGS), which produces blur scores with a fixed range of (0,1). A blurred image is first reblurred using a Gaussian low-pass filter, producing a heavily blurred image. With this reblurred image as reference, a local blur map is generated by computing the gradient similarity. Finally, visual saliency is employed in the pooling to adapt to the characteristics of the human visual system (HVS). The proposed metric features fixed range, fast computation and better consistency with the HVS. Experiments demonstrate its advantages.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2015EDL8041/_p
Copy
@ARTICLE{e98-d_8_1613,
author={Peipei ZHAO, Leida LI, Hao CAI, },
journal={IEICE TRANSACTIONS on Information},
title={Saliency Guided Gradient Similarity for Fast Perceptual Blur Assessment},
year={2015},
volume={E98-D},
number={8},
pages={1613-1616},
abstract={Blur is one of the most common distortion type and greatly impacts image quality. Most existing no-reference (NR) image blur metrics produce scores without a fixed range, so it is hard to judge the extent of blur directly. This letter presents a NR perceptual blur metric using Saliency Guided Gradient Similarity (SGGS), which produces blur scores with a fixed range of (0,1). A blurred image is first reblurred using a Gaussian low-pass filter, producing a heavily blurred image. With this reblurred image as reference, a local blur map is generated by computing the gradient similarity. Finally, visual saliency is employed in the pooling to adapt to the characteristics of the human visual system (HVS). The proposed metric features fixed range, fast computation and better consistency with the HVS. Experiments demonstrate its advantages.},
keywords={},
doi={10.1587/transinf.2015EDL8041},
ISSN={1745-1361},
month={August},}
Copy
TY - JOUR
TI - Saliency Guided Gradient Similarity for Fast Perceptual Blur Assessment
T2 - IEICE TRANSACTIONS on Information
SP - 1613
EP - 1616
AU - Peipei ZHAO
AU - Leida LI
AU - Hao CAI
PY - 2015
DO - 10.1587/transinf.2015EDL8041
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E98-D
IS - 8
JA - IEICE TRANSACTIONS on Information
Y1 - August 2015
AB - Blur is one of the most common distortion type and greatly impacts image quality. Most existing no-reference (NR) image blur metrics produce scores without a fixed range, so it is hard to judge the extent of blur directly. This letter presents a NR perceptual blur metric using Saliency Guided Gradient Similarity (SGGS), which produces blur scores with a fixed range of (0,1). A blurred image is first reblurred using a Gaussian low-pass filter, producing a heavily blurred image. With this reblurred image as reference, a local blur map is generated by computing the gradient similarity. Finally, visual saliency is employed in the pooling to adapt to the characteristics of the human visual system (HVS). The proposed metric features fixed range, fast computation and better consistency with the HVS. Experiments demonstrate its advantages.
ER -