DOI:
https://doi.org/10.64539/sjer.v1i2.2025.25Keywords:
Noise Estimation, Markov Processes, Maximum Likelihood Estimation, Convolutional Neural Networks, Image RestorationAbstract
The assessment of complex noise in textured images requires a method which uses both Markov processes together with Maximum Likelihood Estimation and Convolutional Neural Networks. The evaluation of noise through traditional methods does not deliver acceptable results during preservation of image characteristics in areas with challenging texture patterns. Through Maximum Likelihood Estimation (MLE) probabilistic refinement together with Convolutional Neural Networks (CNNs) features the proposed model applies Markov processes to maintain spatial dependencies that provide accurate denoising with protected image quality. Using CNN-based denoising together with Gaussian filtering creates superior outcomes for imaging perception than individual methods during Edge Preservation Index (EPI) and Structural Similarity Index (SSIM) and Peak Signal-to-Noise Ratio (PSNR) assessment. The experimental results show a 24.85 dB PSNR value together with 0.92 SSIM integrity and EPI quality of 0.85 for effective hybrid model noise reduction. The research utilizes Markov processes and MLE together with Convolutional Neural Networks to develop an all-encompassing approach for cleaning texturized complex images which could serve multiple image types including those from medical contexts and satellites and digital photographs.
References
[1] J. Su, B. Xu, and H. Yin, “A survey of deep learning approaches to image restoration,” Neurocomputing, vol. 487, pp. 46–65, May 2022, doi: 10.1016/j.neucom.2022.02.046.
[2] K. Li, W. Zhou, H. Li, and M. A. Anastasio, “A Hybrid Approach for Approximating the Ideal Observer for Joint Signal Detection and Estimation Tasks by Use of Supervised Learning and Markov-Chain Monte Carlo Methods,” IEEE Trans Med Imaging, vol. 41, no. 5, pp. 1114–1124, May 2022, doi: 10.1109/TMI.2021.3135147.
[3] S. Sun, W. Ren, Z. Liu, H. Park, R. Wang, and X. Cao, “EnsIR: An Ensemble Algorithm for Image Restoration via Gaussian Mixture Models,” in 38th Conference on Neural Information Processing Systems (NeurIPS 2024)., 2024, pp. 1–31.
[4] K. Gong, C. Catana, J. Qi, and Q. Li, “PET Image Reconstruction Using Deep Image Prior,” IEEE Trans Med Imaging, vol. 38, no. 7, pp. 1655–1665, Jul. 2019, doi: 10.1109/TMI.2018.2888491.
[5] O. Özdenizci and R. Legenstein, “Restoring Vision in Adverse Weather Conditions With Patch-Based Denoising Diffusion Models,” IEEE Trans Pattern Anal Mach Intell, vol. 45, no. 8, pp. 10346–10357, Aug. 2023, doi: 10.1109/TPAMI.2023.3238179.
[6] S. España-Boquera, M. J. Castro-Bleda, J. Gorbe-Moya, and F. Zamora-Martinez, “Improving Offline Handwritten Text Recognition with Hybrid HMM/ANN Models,” IEEE Trans Pattern Anal Mach Intell, vol. 33, no. 4, pp. 767–779, Apr. 2011, doi: 10.1109/TPAMI.2010.141.
[7] A. E. Ilesanmi and T. O. Ilesanmi, “Methods for image denoising using convolutional neural network: a review,” Complex & Intelligent Systems, vol. 7, no. 5, pp. 2179–2198, Oct. 2021, doi: 10.1007/s40747-021-00428-4.
[8] J. Peng, L. Li, and Y. Y. Tang, “Maximum Likelihood Estimation-Based Joint Sparse Representation for the Classification of Hyperspectral Remote Sensing Images,” IEEE Trans Neural Netw Learn Syst, vol. 30, no. 6, pp. 1790–1802, Jun. 2019, doi: 10.1109/TNNLS.2018.2874432.
[9] S. Zhang, J. Liu, B. Hu, and Z. Mao, “GH-DDM: the generalized hybrid denoising diffusion model for medical image generation,” Multimed Syst, vol. 29, no. 3, pp. 1335–1345, Jun. 2023, doi: 10.1007/s00530-023-01059-0.
[10] T. Küstner, K. Hammernik, D. Rueckert, T. Hepp, and S. Gatidis, “Predictive uncertainty in deep learning–based MR image reconstruction using deep ensembles: Evaluation on the fastMRI data set,” Magn Reson Med, vol. 92, no. 1, pp. 289–302, Jul. 2024, doi: 10.1002/mrm.30030.
[11] C. Tan, S. Lv, F. Dong, and M. Takei, “Image Reconstruction Based on Convolutional Neural Network for Electrical Resistance Tomography,” IEEE Sens J, vol. 19, no. 1, pp. 196–204, Jan. 2019, doi: 10.1109/JSEN.2018.2876411.
[12] W. Dong, P. Wang, W. Yin, G. Shi, F. Wu, and X. Lu, “Denoising Prior Driven Deep Neural Network for Image Restoration,” IEEE Trans Pattern Anal Mach Intell, vol. 41, no. 10, pp. 2305–2318, Oct. 2019, doi: 10.1109/TPAMI.2018.2873610.
[13] X. Song, Y. Xu, and F. Dong, “A hybrid regularization method combining Tikhonov with total variation for electrical resistance tomography,” Flow Measurement and Instrumentation, vol. 46, pp. 268–275, Dec. 2015, doi: 10.1016/j.flowmeasinst.2015.07.001.
[14] D.-G. Kim, Y. Ali, M. A. Farooq, A. Mushtaq, M. A. A. Rehman, and Z. H. Shamsi, “Hybrid Deep Learning Framework for Reduction of Mixed Noise via Low Rank Noise Estimation,” IEEE Access, vol. 10, pp. 46738–46752, 2022, doi: 10.1109/ACCESS.2022.3170490.
[15] K. Sohn, H. Lee, and X. Yan, “Learning Structured Output Representation using Deep Conditional Generative Models,” in Advances in Neural Information Processing Systems 28 (NIPS 2015), 2015, pp. 1–9.
[16] Y. Guo, X. Wu, C. Qing, C. Su, Q. Yang, and Z. Wang, “Blind Restoration of Images Distorted by Atmospheric Turbulence Based on Deep Transfer Learning,” Photonics, vol. 9, no. 8, p. 582, Aug. 2022, doi: 10.3390/photonics9080582.
[17] D. Hwang et al., “Improving the Accuracy of Simultaneously Reconstructed Activity and Attenuation Maps Using Deep Learning,” Journal of Nuclear Medicine, vol. 59, no. 10, pp. 1624–1629, Oct. 2018, doi: 10.2967/jnumed.117.202317.
[18] X. Fu, H. Liang, and S. Jia, “Mixed Noise-Oriented Hyperspectral and Multispectral Image Fusion,” IEEE Transactions on Geoscience and Remote Sensing, vol. 61, pp. 1–16, 2023, doi: 10.1109/TGRS.2023.3323480.
[19] V. Rajesh and P. Sivakumar, “Deep learning based denoising and enhancement of satellite images using DA-CNN and ORHE techniques,” Earth Sci Inform, vol. 18, no. 1, p. 153, Jan. 2025, doi: 10.1007/s12145-024-01578-y.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Mst Jannatul Kobra, Md Owahedur Rahman, Arman Mohammad Nakib
This work is licensed under a Creative Commons Attribution 4.0 International License.
