Generation of Noisy Images in Extremely Low-Light Environments

QIN Jiahao; QIN Pinle; CHAI Rui; CHEN Zuojun; GAO Yipeng; Wang Bao

Journal of North University of China(Natural Science Edition) > 2024 > 45(5): 608-613.

QIN Jiahao, QIN Pinle, CHAI Rui, CHEN Zuojun, GAO Yipeng, Wang Bao. Generation of Noisy Images in Extremely Low-Light Environments[J]. Journal of North University of China(Natural Science Edition), 2024, 45(5): 608-613.

Citation:

PDF (1423 KB)

Generation of Noisy Images in Extremely Low-Light Environments

1. School of Computer Science and Technology, North University of China;
2. Shanxi Hexin Jiye Technology Co., Ltd.;
3. Shanxi Coking Coal Huajin Coking Coal Co., Ltd.Shaqu No.2 Coal Mine

More Information

Received Date: December 26, 2023

Graphical Abstract

Abstract

Abstract

Due to the complexity of noise in low light environments, it is very difficult to establish noise models. In this regard, this paper proposed a noise image generation method based on diffusion model.Firstly, this article introduced various types of noise into clean images using the forward process of diffusion models based on prior knowledge. Then, a conditional diffusion model was used to input it together with the clean image into the network. Finally, the reverse process of the cold diffusion model was used to iteratively generate noisy images. The experimental results on a low light dataset show that compared with other algorithms, the noise images generated by our algorithm have an objective KL divergence value of 0. 068, which is 0. 001 lower than the existing best methods. The subjective quality is higher, and it is closest to images in low light environments. The established noise model is the most accurate. This method successfully established a high-quality noise model in low light environments, which provided a new approach for noise modeling in low light environments, and the diffusion model was used to the field of low light noise modeling.
- low light imaging,
- diffusion model,
- noisy image generation,
- image process,
- neural network,
- deep learning

FullText(HTML)

References (21)

References

[1]	MCLEAN I S, WALKER G. Electronic imaging in astronomy:Detectors and instrumentation[J]. Physics Today, 1998, 51(6):66.
[2]	陈钱.先进夜视成像技术发展探讨[J].红外与激光工程,2022,51(2):1-8.CHEN Qian. Discussions on the development of advanced night vision imaging technology[J]. Infrared and Laser Engineering, 2022,51(2):1-8.(in Chinese)
[3]	CHEN C, CHEN Q F, XU J, et al. Learning to see in the dark[C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018:3291-3300.
[4]	CHEN C, CHEN Q F, DO M, et al. Seeing motion in the dark[C]//IEEE/CVF International Conference on Computer Vision(ICCV), 2019:3184-3193.
[5]	DABOV K, FOI A, KATKOVNIK V, et al. Image denoising by sparse 3-d transform-domain collaborative filtering[J]. IEEE Transactions on Image Processing, 2007, 16(8):2080-2095.
[6]	MAGGIONI M, BORACCHI G, FOI A, et al.Video denoising, deblocking, and enhancement through separable 4-d nonlocal spatiotemporal transforms[J]. IEEE Transactions on Image Processing,2012, 21(9):3952-3966.
[7]	WEI K X, FU Y, ZHENG Y Q, et al. Physics-based noise modeling for extreme low-light photography[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(11):8520-8537.
[8]	MONAKHOVA K, RICHTER S R, WALLER L,et al. Dancing under the stars:video denoising in starlight[C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR), 2022:16220-16230.
[9]	GULRAJANI I, AHMED F, ARJOVSKY M, et al.Improved training of wasserstein gans[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems, 2017:5769-5779.
[10]	SAHARIA C, HO J, CHAN W, et al. Image superresolution via iterative refinement[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2022, 45(4):4713-4726.
[11]	OSHER S, BURGER M, GOLDFARB D, et al. An iterative regularization method for total variation-based image restoration[J]. Multiscale Modeling&Simulation, 2005, 4(2):460-489.
[12]	RUDIN L I, OSHER S, FATEMI E. Nonlinear total variation based noise removal algorithms[J]. Physica D:Nonlinear Phenomena, 1992, 60(1/2/3/4):259-268.
[13]	陈作钧,秦品乐,柴锐,等.微光场景下的时空域EMCCD视频降噪算法[J].中北大学学报(自然科学版),2023, 44(2):168-175.CHEN Zuojun, QIN Pinle, CHAI Rui, et al. Spatiotemporal domain EMCCD video noise eduction algorithm in low light scene[J]. Journal of North University of China(Natural Science Edition), 2023,44(2):168-175.(in Chinese)
[14]	ZHANG K, ZUO W M, CHEN Y J, et al. Beyond a gaussian denoiser:Residual learning of deep cnn for image denoising[J]. IEEE Transactions on Image Processing, 2017, 26(7):3142-3155.
[15]	FOI A, TRIMECHE M, KATKOVNIK V, et al.Practical poissonian-gaussian noise modeling and fitting for single-image raw-data[J]. IEEE Transactions on Image Processing, 2008, 17(10):1737-1754.
[16]	GODARD C, MATZEN K, UYTTENDAELE M.Deep burst denoising[C]//European Conference on Computer vision(ECCV), 2018:538-554.
[17]	HO J, JAIN A, ABBEEL P. Denoising diffusion probabilistic models[J]. Advances in Neural Information Processing Systems, 2020, 33:6840-6851.
[18]	HO J, SALIMANS T. Classifier-free diffusion guidance[DB/OL].(2022-07-26)[2023-12-27]. http://arxiv. org/abs/2207. 12598.
[19]	BANSAL A, BORGNIA E, CHU H M, et al. Cold diffusion:Inverting arbitrary image transforms without noise[DB/OL].(2022-08-19)[2023-12-27]. http://arxiv. org/abs/2208. 09392.
[20]	CHANG K C, WANG R, LIN H J, et al. Learning camera-aware noise models[C]//Computer Vision ECCV, 2020:343-358.
[21]	ABDELHAMED A, BRUBAKER M A, BROWN M S. Noise flow:Noise modeling with conditional normalizing flows[C]//IEEE/CVF International Conference on Computer Vision, 2019:3165-3173.