Abstract
In dark or poorly lit environments, it is often difficult for the naked eye to distinguish low-light-level images because of low brightness, low contrast and noise, and it is difficult to perform subsequent image processing (such as video surveillance and target detection). To solve these problems, this paper proposes a low-light-level image enhancement algorithm based on deep learning. First, the low-light-level image is segmented into several super-pixels, and the noise level of each super-pixel is estimated by the ratio of the local standard deviation to the local gradient. Then, the image is inverted and smoothed by a BM3D filter, and the structural filter adaptive method is used to obtain complete images without noise but with the correct texture. Finally, the noise-free image and texture-complete images are applied to the integrated network, which can not only enhance the contrast but also effectively prevent the over-enhancement of the contrast. The experimental results show that this method is superior to traditional methods in terms of both subjective and objective evaluation, and the peak signal–noise ratio and improved structural similarity are 31.64 dB and 91.2%, respectively.
Similar content being viewed by others
References
Talaulikar, A.S., Sanathanan, S., Modi, C.N.: An enhanced approach for detecting helmet on motorcyclists using image processing and machine learning techniques. Advanced Computing and Communication Technologies (2019)
Yan, C.: Several contrast enhancement algorithms for low-light images [J]. J. Jingdezhen Univ. 6, 10–12 (2013)
Bo, P., Yiming, W., Pengbo, et al.: Research and implementation of low illumination image enhancement algorithm. Comput. Appl. (6) (2007)
Lili, D., Chang, D., Wenhai, X.: Two improved methods of image enhancement based on histogram equalization [J]. J. Electron. 46(10), 65–73 (2018)
Xueming, L.: Image enhancement algorithm based on Retinex theory [J]. Comput. Appl. Res. 22(2), 235–237 (2005)
Li, L., Wang, R., Wang, W., et al. A low-light image enhancement method for both de-noising and contrast enlarging[C]//2015 IEEE International Conference on Image Processing (ICIP). IEEE, September 27–30, 2015, Quebec City, Canada
Yandong, L., Zongbo, H., Leihang. Review of convolutional neural networks [J].Comput. Appl. 36(9), 2508–2515 (2016)
Hou, J.-C., Wang, S.-S., Lai, Y.-H., et al.: Audio-visual speech enhancement based on multimodal deep convolutional neural network (2017). arXiv preprint arXiv:1804.03641
Chao, C., Chao, C. Application of improved single-scale Retinex algorithm in image enhancement [J]. Comput. Appl. Softw. 30(4), 55–57 (2013)
Watanabe, T., Kuwahara, Y,. Kojima, A., Kurosawa, T.: Improvement of color quality with modified linear multiscale retinex. Proc. SPIE 5008, Color imaging VIII: processing, hardcopy, and applications, 13 January 2003 (2003). https://doi.org/10.1117/12.472030
Chun, H., Guoqin, S.: Algorithmic implementation based on MSRCR theory [J]. Intell. Comput. Appl. 7(3), 114–116 (2017)
Fu, X., Zeng, D., Yue, H. et al. A weighted variational model for simultaneous reflectance and illumination estimation[C]//computer vision & pattern recognition. Las Vegas, Nevada, USA, Jun 26, 2016–Jul 1, 2016
Guo, X., Li, Y., Ling, H.: LIME: low-light image enhancement via illumination map estimation[J]. IEEE Trans. Image Process. 26(2), 982–993 (2017)
Dong, C., Loy, C.C., He, K., et al.: Image super-resolution using deep convolutional networks[J]. IEEE Trans. Pattern Anal. Mach. Intell. 38(2), 295–307 (2014)
Guoliang, Y., Nan, X., Fang, L., et al. Research on the deep learning classification method of non-linear activation function [J]. J. Jiangxi Univ. Technol. 39(193)(3), 79–86 (2018)
Yandong, L., Zongbo, H., Leihang. A review of Convolutional neural network [J]. Comput. Appl. 36(9), 2508–2515 (2016)
Djurović, I.: BM3D filter in salt-and-pepper noise removal[J]. Eurasip J. Image Video Process. 2016(1), 13 (2016)
Huanqing, Q.: Design of lattice structure filter [J]. Dig. Technol. Appl. 5, 181–185 (2014)
Abdullah, H.R., Aleawy Algazi, N.A., Bayes estimators for the shape parameter of pareto type i distribution under generalized square error loss function[J]. Soc. Sci. Electron. Publ. 20–32 (2014)
Xiaoqiao, H., Junsheng, S., Jian, Y., et al. Evaluation of color image quality based on mean square error and peak signal-to-noise ratio [J]. J. Photon. 36(Sup1)
Tong, Y.B., Zhang, Q.S., Yun-Ping, Q.I.: Image quality assessing by combining PSNR with SSIM[J]. J. Image Graph. 11(12), 1758–1763 (2006)
Bergh, M.V.D, Boix, X., Roig, G., et al. SEEDS: superpixels extracted via energy-driven sampling[C]//European conference on computer vision. Springer, Berlin (2012)
Cai, B., Xu, X., Jia, K., et al. DehazeNet: an end-to-end system for single image haze removal.[J]. IEEE Trans. Image Process. 25(11), 5187–5198 (2016)
Ren, W., Si, L., Hua, Z., et al. Single image dehazing via multi-scale convolutional neural networks[M]//computer vision—ECCV 2016. 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016
Zhang, Z, He, T, Zhang, H, et al. Bag of freebies for training object detection neural networks (2019). arXiv preprint arXiv:1902.04103
He, K., Zhang, X., Ren, S., et al. Delving deep into rectifiers: surpassing human-level performance on imagenet classification (2015). arXiv preprint arXiv:1502.01852
Weihe, Z., Junwei, Yang, Yuxiang, X.: Bidirectional equalization of dynamic histogram for image enhancement [J]. Stereol. Image Anal. China 2, 129–139 (2014)
Hongqiang, M., Shiping, M., Yueli, X., et al. Low illumination image enhancement based on deep convolution neural network [J]. J. Opt. 39(2) (2019)
Li, T., Zhu, C., Xiang, G., et al. LLCNN: a convolutional neural network for low-light image enhancement[c]//visual communications & image processing. Taichung, Taiwan, December 09 (2018)
Acknowledgements
This work was supported by the Scientific Research Program funded by Shaanxi Science and Technology Department (2019GY-022, 2019GY-066), the National Natural Science Foundation of China (61671362), and the Science and Technology Program of Weiyang District Science and Technology Department (201923).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Wang, P., Wu, J., Wang, H. et al. Low-light-level image enhancement algorithm based on integrated networks. Multimedia Systems 28, 2015–2025 (2022). https://doi.org/10.1007/s00530-020-00671-8
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00530-020-00671-8