Low-light image enhancement based on multi-illumination estimation and multi-scale fusion

Abstract

To improve image quality under low illumination conditions, a novel low-light image enhancement method is proposed in this paper based on multi-illumination estimation and multi-scale fusion (MIMS). Firstly, the illumination is processed by contrast-limited adaptive histogram equalization (CLAHE), adaptive complementary gamma function (ACG), and adaptive detail preserving S-curve (ADPS), respectively, to obtain three components. Then, the fusion-relevant features, exposure, and color contrast are selected as the weight maps. Subsequently, these components and weight maps are fused through multi-scale to generate enhanced illumination. Finally, the enhanced images are obtained by multiplying the enhanced illumination and reflectance. Compared with existing approaches, this proposed method achieves an average increase of 0.81% and 2.89% in the structural similarity index measurement (SSIM) and peak signal-to-noise ratio (PSNR), and a decrease of 6.17% and 32.61% in the natural image quality evaluator (NIQE) and gradient magnitude similarity deviation (GMSD), respectively.

References

[1] WangW C, WuX J, GaoZ R, et al.. An experiment-based review of low-light image enhancement methods. IEEE access, 2020, 8: 87884-87917 J]

[2] ZhangQ S, OuyangZ W, RenZ L, et al.. CUCN: continuously updated connection network for low-light image enhancement. Journal of electronic imaging, 2023, 32(3): 033010 J]

[3] SaadN H, IsaN A M, SalehH M. Nonlinear exposure intensity based modification histogram equalization for non-uniform illumination image enhancement. IEEE access, 2021, 9: 93033-93061 J]

[4] SinghN, BhandariA K. Principal component analysis-based low-light image enhancement using reflection model. IEEE transactions on instrumentation and measurement, 2021, 70: 1-10[J]

[5] YiX, MinC B, ShaoM C, et al.. Low-light image enhancement via regularized Gaussian fields model. Neural processing letters, 2023, 55: 12017-12037 J]

[6] LangY Z, QianY S, KongX Y, et al.. Effective enhancement method of low-light-level images based on the guided filter and multi-scale fusion. Journal of the optical society of America A, optics, image science, and vision, 2022, 40(1): 1-9 J]

[7] DemirY, KaplanN H. Low-light image enhancement based on sharpening-smoothing image filter. Digital signal process, 2023, 138: 104054 J]

[8] MertensT, KautzJ, ReethV F. Exposure fusion. 15th Pacific Conference on Computer Graphics and Applications (PG’07), October 29–November 2, 2007, Maui, HI, USA, 2007, New York, IEEE: 382-390[C]

[9] FuX Y, ZengD L, HuangY, et al.. A fusion-based enhancing method for weakly illuminated images. Signal processing, 2016, 129: 82-96 J]

[10] MarquesT P, BranzanA A. L²UWE: a framework for the efficient enhancement of low-light underwater images using local contrast and multi-scale fusion. IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, July 14–19, 2020, Seattle, WA, USA, 2020, New York, IEEE: 2286-2295[C]

[11] ChenJ, WuY. Improved color balance and fusion for enhancement of underwater image. 2022 IEEE 8th International Conference on Computer and Communications (ICCC), December 9–12, 2022, Chengdu, China, 2023, New York, IEEE: 1964-1969[C]

[12] ZhouX Y, GuoJ C, LiuG, et al.. A fusion-based and multi-layer method for low light image enhancement. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), June 4–10, 2023, Rhodes Island, Greece, 2023, New York, IEEE: 1-5[C]

[13] ZhangE, GuoL, GuoJ, et al.. A low-brightness image enhancement algorithm based on multi-scale fusion. Applied sciences, 2023, 13(8): 10230 J]

[14] HanY, ZhangW, HeW. Low-light image enhancement based on simulated multi-exposure fusion. Journal of physics: conference series, 2023, 2478(6): 062022[J]

[15] HeZ, MoH, XiaoY, et al.. Multi-scale fusion for image enhancement in shield tunneling: a combined MSRCR and CLAHE approach. Measurement science and technology, 2024, 35(5): 056112 J]

[16] ZhongW, LinJ, MaL, et al.. Deep multi-illumination fusion for low-light image enhancement. Pattern recognition and computer vision, 2021, 13021: 140-150 J]

[17] BurtP J, KolczynskiR J. Enhanced image capture through fusion. 4th International Conference on Computer Vision, May 11–14, 1993, Berlin, Germany, 2002, New York, IEEE[C]

[18] PariharA S, SinghK, RohillaH, et al.. Fusion-based simultaneous estimation of reflectance and illumination for low-light image enhancement. IET image processing, 2020, 15: 1410-1423 J]

[19] Retinex image processing[EB/OL]. (2001) [2024-02-13]. https://dragon.larc.nasa.gov/retinex/pao/news.

[20] GuoX J, LiY, LingH B. LIME: low-light image enhancement via illumination map estimation. IEEE transactions on image processing, 2017, 26(2): 982-993 J]

[21] CaiJ R, GuS H, ZhangL. Learning a deep single image contrast enhancer from multi-exposure images. IEEE transactions on image processing, 2018, 27(4): 2049-2062 J]

[22] LeeC L, LeeC, KimC S. Contrast enhancement based on layered difference representation of 2D histograms. 19th IEEE International Conference on Image Processing, September 30-October 3, 2012, Orlando, FL, USA, 2013, New York, IEEE: 965-968[C]

RIGHTS & PERMISSIONS

Tianjin University of Technology