Super resolution (SR) image reconstruction has attracted increasing attention these years and many SR image reconstruction algorithms have been proposed for restoring a high-resolution image from one or multiple low-resolution images. However, how to objectively evaluate the quality of SR reconstructed images remains an open problem. Although a great number of image quality metrics have been proposed, they are quite limited to evaluate the quality of SR reconstructed images. Inspired by this, this paper presents a blind quality index for SR reconstructed images using first- and second-order structural degradation. First, the SR reconstructed image is decomposed into multi-order derivative magnitude maps, which are effective for first- and second-order structural representation. Then, log-energy based features are extracted on these multi-order derivative magnitude maps in the frequency domain. Finally, support vector regression is used to learn the quality model for SR reconstructed images. The results of extensive experiments that were conducted on one public database demonstrate the superior performance of the proposed method over the existing quality metrics. Moreover, the proposed method is less dependent on the number of training images and has low computational cost.
Jiansheng QIAN
China University of Mining and Technology
Bo HU
China University of Mining and Technology
Lijuan TANG
Jiangsu Vocational College of Business
Jianying ZHANG
China University of Mining and Technology
Song LIANG
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
Jiansheng QIAN, Bo HU, Lijuan TANG, Jianying ZHANG, Song LIANG, "Blind Quality Index for Super Resolution Reconstructed Images Using First- and Second-Order Structural Degradation" in IEICE TRANSACTIONS on Fundamentals,
vol. E102-A, no. 11, pp. 1533-1541, November 2019, doi: 10.1587/transfun.E102.A.1533.
Abstract: Super resolution (SR) image reconstruction has attracted increasing attention these years and many SR image reconstruction algorithms have been proposed for restoring a high-resolution image from one or multiple low-resolution images. However, how to objectively evaluate the quality of SR reconstructed images remains an open problem. Although a great number of image quality metrics have been proposed, they are quite limited to evaluate the quality of SR reconstructed images. Inspired by this, this paper presents a blind quality index for SR reconstructed images using first- and second-order structural degradation. First, the SR reconstructed image is decomposed into multi-order derivative magnitude maps, which are effective for first- and second-order structural representation. Then, log-energy based features are extracted on these multi-order derivative magnitude maps in the frequency domain. Finally, support vector regression is used to learn the quality model for SR reconstructed images. The results of extensive experiments that were conducted on one public database demonstrate the superior performance of the proposed method over the existing quality metrics. Moreover, the proposed method is less dependent on the number of training images and has low computational cost.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.E102.A.1533/_p
Copy
@ARTICLE{e102-a_11_1533,
author={Jiansheng QIAN, Bo HU, Lijuan TANG, Jianying ZHANG, Song LIANG, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Blind Quality Index for Super Resolution Reconstructed Images Using First- and Second-Order Structural Degradation},
year={2019},
volume={E102-A},
number={11},
pages={1533-1541},
abstract={Super resolution (SR) image reconstruction has attracted increasing attention these years and many SR image reconstruction algorithms have been proposed for restoring a high-resolution image from one or multiple low-resolution images. However, how to objectively evaluate the quality of SR reconstructed images remains an open problem. Although a great number of image quality metrics have been proposed, they are quite limited to evaluate the quality of SR reconstructed images. Inspired by this, this paper presents a blind quality index for SR reconstructed images using first- and second-order structural degradation. First, the SR reconstructed image is decomposed into multi-order derivative magnitude maps, which are effective for first- and second-order structural representation. Then, log-energy based features are extracted on these multi-order derivative magnitude maps in the frequency domain. Finally, support vector regression is used to learn the quality model for SR reconstructed images. The results of extensive experiments that were conducted on one public database demonstrate the superior performance of the proposed method over the existing quality metrics. Moreover, the proposed method is less dependent on the number of training images and has low computational cost.},
keywords={},
doi={10.1587/transfun.E102.A.1533},
ISSN={1745-1337},
month={November},}
Copy
TY - JOUR
TI - Blind Quality Index for Super Resolution Reconstructed Images Using First- and Second-Order Structural Degradation
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1533
EP - 1541
AU - Jiansheng QIAN
AU - Bo HU
AU - Lijuan TANG
AU - Jianying ZHANG
AU - Song LIANG
PY - 2019
DO - 10.1587/transfun.E102.A.1533
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E102-A
IS - 11
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - November 2019
AB - Super resolution (SR) image reconstruction has attracted increasing attention these years and many SR image reconstruction algorithms have been proposed for restoring a high-resolution image from one or multiple low-resolution images. However, how to objectively evaluate the quality of SR reconstructed images remains an open problem. Although a great number of image quality metrics have been proposed, they are quite limited to evaluate the quality of SR reconstructed images. Inspired by this, this paper presents a blind quality index for SR reconstructed images using first- and second-order structural degradation. First, the SR reconstructed image is decomposed into multi-order derivative magnitude maps, which are effective for first- and second-order structural representation. Then, log-energy based features are extracted on these multi-order derivative magnitude maps in the frequency domain. Finally, support vector regression is used to learn the quality model for SR reconstructed images. The results of extensive experiments that were conducted on one public database demonstrate the superior performance of the proposed method over the existing quality metrics. Moreover, the proposed method is less dependent on the number of training images and has low computational cost.
ER -