CNN Based No-Reference HDR Image Quality Assessment

FAN Kefeng; LIANG Jiyun; LI Fei; QIU Puye

doi:10.1049/cje.2021.01.008

Volume 30 Issue 2

Apr. 2021

Turn off MathJax

Article Contents

Article Navigation > Chinese Journal of Electronics > 2021 > 30(2): 282-288

FAN Kefeng, LIANG Jiyun, LI Fei, et al., “CNN Based No-Reference HDR Image Quality Assessment,” Chinese Journal of Electronics, vol. 30, no. 2, pp. 282-288, 2021, doi: 10.1049/cje.2021.01.008

Citation:

FAN Kefeng, LIANG Jiyun, LI Fei, et al., “CNN Based No-Reference HDR Image Quality Assessment,” Chinese Journal of Electronics, vol. 30, no. 2, pp. 282-288, 2021, doi: 10.1049/cje.2021.01.008

FAN Kefeng, LIANG Jiyun, LI Fei, et al., “CNN Based No-Reference HDR Image Quality Assessment,” Chinese Journal of Electronics, vol. 30, no. 2, pp. 282-288, 2021, doi: 10.1049/cje.2021.01.008

Citation:

FAN Kefeng, LIANG Jiyun, LI Fei, et al., “CNN Based No-Reference HDR Image Quality Assessment,” Chinese Journal of Electronics, vol. 30, no. 2, pp. 282-288, 2021, doi: 10.1049/cje.2021.01.008

PDF( 1460 KB)

CNN Based No-Reference HDR Image Quality Assessment

doi: 10.1049/cje.2021.01.008

FAN Kefeng^{1, 2
,
,},
LIANG Jiyun^{1, 2
,},
LI Fei^3
,,
QIU Puye^1
,

1.
Digital Technology Research Center, China Electronics Standardization Institute, Beijing 100007, China
2.
School of Electronic Engineering and Automation, Guilin University of Electronic Technology, Guilin 541004, China
3.
Guangzhou SequoiaDB Co., Guangzhou 510006, China

Funds:

The National Key Research and Development Program of China 2019YFB1405503

2019 Public Service Platform of Industrial Technology Foundation of MIIT 2019-00895-2-1

More Information

Author Bio:
LIANG Jiyun   was born in November 1994, Master of engineering. His research interests include video coding and image processing. (E-mail: 18589837505@163.com)

LI Fei   was born in July 1985, senior engineer.Her research interests include cloud computing and database, etc. (E-mail: fiona2022@163.com)

QIU Puye    was born in August 1994, Master of Science.His research interests include video coding and technical system management. (E-mail: qiupy@cesi.cn)
Corresponding author: FAN Kefeng (corresponding author) was born in December 1978, Ph.D., post-doctorate, researcher-level senior engineer. His research interests include intelligent information processing and cyber security. (E-mail: fankf@126.com)
Received Date: 2020-05-06
Accepted Date: 2020-06-09
Publish Date: 2021-03-01

Abstract

Abstract

Motivated by the problems of nonuniversality and over-reliance on the original reference image in High dynamic range (HDR) Image quality assessment (IQA), a convolutional neural network-based algorithm for no-reference HDR image quality assessment is proposed. The Salience detection by self-resemblance (SDSR) algorithm which extracts the salient regions of the HDR image, is used to simulate the human visual attention mechanism. Then a visual quality perception network for training quality prediction models is designed according to the visual characteristics of luminance and contrast sensitivity. And this network consists of an Error estimation network (Error-net), a Perceptual resistance network (PR-net) and a mixing function. The experimental results indicate that the method proposed has high consistency with subjective perception, and the value of assessment metrics Spearman rank-order correlation coefficient (SROCC), Pearson product-moment correlation coefficient (PLCC) and Root mean square error (RMSE) correspondingly reaches 0.941, 0.910 and 8.176 as well. It is comparable with classic full-reference HDR IQA methods.
- High dynamic range image,
- Image quality assessment,
- Convolutional neural network,
- Human visual system

FullText(HTML)

References(16)

References

[1]	Artusi A, Richter T, Ebrahimi T, et al., "High dynamic range imaging technology", IEEE Signal Processing Magazine, 34(5): 165–172, 2017. doi: 10.1109/MSP.2017.2716957
[2]	G. Yue, C. Hou and T. Zhou, "Blind quality assessment of tone-mapped images considering colorfulness, naturalness, and structure", in IEEE Transactions on Industrial Electronics, Vol. 66, No. 5, pp. 3784–3793, 2019. doi: 10.1109/TIE.2018.2851984
[3]	M. Zhao, L. Shen, M. Jiang, et al., "A novel no-reference quality assessment model of tone-mapped HDR image", 2019 IEEE International Conference on Image Processing (ICIP), Taipei, China, pp. 3202–3206, 2019.
[4]	Z. Cui, Z. Gan, G. Tang, et al., "Image signature based mean square error for image quality assessment", Chinese Journal of Electronics, Vol. 24, No. 4, pp. 755–760, 2015. doi: 10.1049/cje.2015.10.015
[5]	CHEN Yang, LI Dan, ZHANG Jian-qiu, "Blind image quality assessment with complementary color wavelet transform", Acta Electronica Sinica, Vol. 47, No. 4, pp. 775–783, 2019. http://en.cnki.com.cn/Article_en/CJFDTotal-DZXU201904002.htm
[6]	Hanhart P., Bernardo M.V., Pereira M. et al., "Benchmarking of objective quality metrics for HDR image quality assessment", EURASIP Journal on Image and Video Processing, Article No. 39, DOI: 10.1186/s13640-015-0091-4,2015.
[7]	Narwaria M, Mantiuk R, Da Silva M P, et al. "HDR-VDP-2.2: A calibrated method for objective quality prediction of high-dynamic range and standard images", Journal of Electronic Imaging, Vol. 24, No. 1, Article No. 010501, 2015.
[8]	Narwaria M, Da Silva M P and Callet P L, "HDR-VQM: An objective quality measure for high dynamic range video", Signal Processing: Image Communication, Vol. 35, pp. 46–60, 2015. doi: 10.1016/j.image.2015.04.009
[9]	YU Jiaowen, YU Mei, SHAO Hua and JIANG Gangyi, "High dynamic range image quality assessment based on manifold learning", Laser Journal, Vol. 38, No. 4, pp. 90–95, 2017.
[10]	Zou Liangtao, Jiang Gangyi, Yu Mei, et al., "No-reference image quality assessment of high dynamic range image based on tensor domain perceptual features", Journal of Computer-Aided Design & Computer Graphics, Vol. 30, No. 10, DOI: 10.3724/SP.J.1089.2018.17014, 2018.
[11]	Guan F, Jiang G, Song Y, et al., "No-reference high-dynamic-range image quality assessment based on tensor decomposition and manifold learning", Applied Optics, Vol. 57, No. 4, pp. 839–848, 2018. doi: 10.1364/AO.57.000839
[12]	Seo H J and Milanfar P. "Static and space-time visual saliency detection by self-resemblance", Journal of Vision, Vol. 9, No. 12, DOI: 10.1167/9.12.15,2009.
[13]	T. Zhu and L. Karam, "A no-reference objective image quality metric based on perceptually weighted local noise", EURASIP J. Image Video Process., Vol. 2014, No. 1, pp. 1–5, 2014. doi: 10.1186/1687-5281-2014-5
[14]	HDR Photographic Survey "Fairchild dataset", http://rit-mcsl.org/fairchild//HDR.html., 2017-8-17.
[15]	Narwaria M, Perreira Da Silva M, Le Callet P, et al., "Tone mapping-based high-dynamic range image compression: Study of optimization criterion and perceptual quality", Optical Engineering, Vol. 52, No. 10, pp. 102008-1–102008-15, October 2013.
[16]	Zerman E, Valenzise G and Dufaux F, "An extensive performance evaluation of full-reference HDR image quality metrics", Quality & User Experience, Vol. 2, No. 1, Article No. 5, 2017.