Sensor-Calibrated Poisson–Gaussian Modeling and Adaptive Algorithms for RAW Image Denoising
DOI:
https://doi.org/10.54097/td33y502Keywords:
Adaptive algorithm, BM3D, image denoising, NAFNet, photon transfer curve, Poisson–Gaussian model, RAW imagingAbstract
RAW images preserve information close to the camera sensor, but they also contain noise whose strength changes with brightness. This study develops a denoising workflow for monochrome 10-bit RAW video from an industrial camera. The camera noise was first measured with a Photon Transfer Curve (PTC). Two independently recorded grayscale videos produced the fitted relation variance ≈ 0.106185 × mean + 2.001, supporting a Poisson–Gaussian noise model. The calibrated model was then used in a variance-stabilizing transform before BM3D, creating a strong classical baseline. A multi-frame NAFNet system was trained with synthetic noise generated from the PTC and reference images obtained from real camera bursts. Because a single network did not balance noise removal and edge preservation well, the final system combined a clean-image branch with an edge-preserving branch. A light wavelet refinement with automatically selected strength was applied only to flat regions. On 11 fixed development-evaluation clips covering 0.5–250 frames per second, blind BM3D achieved a mean Denoise–Edge Score (DES) of 0.8324, calibrated VST+BM3D achieved 0.8882, and the final P150 system achieved 0.9589 while passing an edge-safety test. The results show that camera measurement and carefully limited processing can improve both classical and learning-based denoising. The study also discusses the limits of its camera-specific dataset and internal evaluation metric.
Downloads
References
[1] Foi, A., Trimeche, M., Katkovnik, V., & Egiazarian, K. (2008). Practical Poissonian–Gaussian noise modeling and fitting for single-image raw-data. IEEE Transactions on Image Processing, 17(10), 1737–1754. https://doi.org/10.1109/TIP.2008.2001399
[2] Dabov, K., Foi, A., Katkovnik, V., & Egiazarian, K. (2007). Image denoising by sparse 3-D transform-domain collaborative filtering. IEEE Transactions on Image Processing, 16(8), 2080–2095. https://doi.org/10.1109/TIP.2007.901238
[3] Mäkitalo, M., & Foi, A. (2013). Optimal inversion of the generalized Anscombe transformation for Poisson–Gaussian noise. IEEE Transactions on Image Processing, 22(1), 91–103. https://doi.org/10.1109/TIP.2012.2202675
[4] Chen, L., Chu, X., Zhang, X., & Sun, J. (2022). Simple baselines for image restoration. In Computer Vision–ECCV 2022 (pp. 17–33). Springer. https://doi.org/10.1007/978-3-031-20071-7_2
[5] Lehtinen, J., Munkberg, J., Hasselgren, J., et al. (2018). Noise2Noise: Learning image restoration without clean data. In Proceedings of the 35th International Conference on Machine Learning (Vol. 80, pp. 2965–2974).
[6] Şendur, L., & Selesnick, I. W. (2002). Bivariate shrinkage functions for wavelet-based denoising exploiting interscale dependency. IEEE Transactions on Signal Processing, 50(11), 2744–2756. https://doi.org/10.1109/TSP.2002.804091
[7] Stein, C. M. (1981). Estimation of the mean of a multivariate normal distribution. The Annals of Statistics, 9(6), 1135–1151. https://doi.org/10.1214/aos/1176345632
[8] Janesick, J. R. (2001). Scientific charge-coupled devices. SPIE Press.
[9] Anscombe, F. J. (1948). The transformation of Poisson, binomial and negative-binomial data. Biometrika, 35(3–4), 246–254. https://doi.org/10.2307/2332343
[10] Hasinoff, S. W., Sharlet, D., Geiss, R., et al. (2016). Burst photography for high dynamic range and low-light imaging on mobile cameras. ACM Transactions on Graphics, 35(6), Article 192. https://doi.org/10.1145/2980179.2980254
[11] Tomasi, C., & Manduchi, R. (1998). Bilateral filtering for gray and color images. In Proceedings of the 6th International Conference on Computer Vision (pp. 839–846). https://doi.org/10.1109/ICCV.1998.710815
[12] Buades, A., Coll, B., & Morel, J.-M. (2005). A non-local algorithm for image denoising. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (pp. 60–65). https://doi.org/10.1109/CVPR.2005.38
[13] Coifman, R. R., & Donoho, D. L. (1995). Translation-invariant de-noising. In A. Antoniadis & G. Oppenheim (Eds.), Wavelets and statistics (pp. 125–150). Springer. https://doi.org/10.1007/978-1-4612-2544-7_9
[14] Wang, Z., Bovik, A. C., Sheikh, H. R., & Simoncelli, E. P. (2004). Image quality assessment: From error visibility to structural similarity. IEEE Transactions on Image Processing, 13(4), 600–612. https://doi.org/10.1109/TIP.2003.819861
[15] Plötz, T., & Roth, S. (2017). Benchmarking denoising algorithms with real photographs. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 1586–1595). https://doi.org/10.1109/CVPR.2017.168
[16] Abdelhamed, A., Lin, S., & Brown, M. S. (2018). A high-quality denoising dataset for smartphone cameras. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 1692–1700). https://doi.org/10.1109/CVPR.2018.00182
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Journal of Mathematical Finance and Risk Management

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.










