Conference Paper2025

Adaptive Denoising Filter Selection for Preprocessing Low-Light Images via Feature-Based Machine Learning Noise Classification System

SM A Sayed Al Banna, Istiaq Ahmed, Ashraful Islam, Rashedur Rahman, Faiza Omar Arpita, Saadia Binte Alam

2025 IEEE 23rd Student Conference on Research and Development (SCOReD)

IEEE, pp. 1-6

DOI: 10.1109/scored68498.2025.11399209URL

Abstract

Image noise degrades computer vision performance in critical fields like medical imaging and autonomous systems. A key challenge is that different noise types require specific treatments, making universal denoising filters ineffective. Non-blind denoising is impractical without prior noise knowledge, while deep learning solutions are often computationally costly for preprocessing. We propose a robust machine learning framework for noise classification based on feature extraction. Our methodology synthesizes a diverse dataset by applying five noise types—Gaussian, Poisson, Salt & Pepper, Quantization, and Periodic—to clean images from the BSDS500 benchmark. From each image, an 8-dimensional feature vector is extracted, capturing distinct statistical, textural, and frequency-domain characteristics. A Random Forest (RF) classifier, optimized via hyperparameter tuning, is trained on this feature set. The model achieves 91% classification accuracy, demonstrating the potency of these features, with analysis revealing the paramount importance of textural descriptors from the Gray-Level Co-occurrence Matrix (GLCM). This work establishes that feature-based machine learning provides a highly accurate, efficient, and interpretable solution for this pivotal preprocessing step in computer vision pipelines.

Keywords

  • periodic noise
  • shot noise
  • quantization noise
  • salt and pepper noise
  • low-light images
  • image denoising
  • gray level co-occurrence matrix
  • textural properties
  • noise classification
  • BSDS500

CCDS Authors

Istiaq Ahmed
Istiaq Ahmed

Istiaq Ahmed

  • Research Assistant, HCI Wing

Published at

CHI, HCII

Ashraful Islam, PhD
Ashraful Islam

Ashraful Islam, PhD

  • Center Director, CCDS
  • Director, HCI Wing

Published at

CHI, UIST, UbiComp, IJCNN, ICIP, HCII, ICDMW, BLP, Pattern Recognition, IEEE Internet of Things Journal, JMIR, JMIR Human Factors, JMIR Formative Research, IEEE Access, Smart Agricultural Technology, Diabetes Research and Clinical Practice, PLOS ONE

Md Rashedur Rahman, D. Eng.
Md. Rashedur Rahman, D. Eng.

Md Rashedur Rahman, D. Eng.

  • Co-Director, MIRA & NEST Wings

  • Assistant Professor

    Department of Computer Science and Engineering

    Independent University, Bangladesh (IUB)

Published at

CHI, HCII, Scientific Reports, Cognitive Computation, Neural Computing and Applications

Saadia Binte Alam, PhD
Saadia Binte Alam, PhD

Saadia Binte Alam, PhD

  • Director, MIRA & WiSE Wings

  • Associate Professor

    Department of Computer Science and Engineering

    Independent University, Bangladesh (IUB)

Published at

HCII, Scientific Reports, Cognitive Computation, Neural Computing and Applications

References

  1. 1.Leonid I. Rudin, Stanley Osher, Emad Fatemi. (1992). Nonlinear total variation based noise removal algorithms. Physica D Nonlinear Phenomena, 60(1-4), 259–268[10.1016/0167-2789(92)90242-f]
  2. 2.D. Martin, C. Fowlkes, D. Tal, J. Malik. (2002). A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics. , 2, 416–423[10.1109/iccv.2001.937655]
  3. 3.A. Buades, B. Coll, J.-M. Morel. (2005). A Non-Local Algorithm for Image Denoising. , 2, 60–65[10.1109/cvpr.2005.38]
  4. 4.T. Huang, G. Yang, G. Tang. (1979). A fast two-dimensional median filtering algorithm. IEEE Transactions on Acoustics Speech and Signal Processing, 27(1), 13–18[10.1109/tassp.1979.1163188]
  5. 5.Linwei Fan, Fan Zhang, Hui Fan, Caiming Zhang. (2019). Brief review of image denoising techniques. Visual Computing for Industry Biomedicine and Art, 2(1), 7[10.1186/s42492-019-0016-7]
  6. 6.A. Pizurica, W. Philips, I. Lemahieu, M. Acheroy. (2003). A versatile wavelet domain noise filtration technique for medical imaging. IEEE Transactions on Medical Imaging, 22(3), 323–331[10.1109/tmi.2003.809588]
  7. 7.Ajay Kumar Boyat, Brijendra Kumar Joshi. (2015). A Review Paper : Noise Models in Digital Image Processing. Signal & Image Processing An International Journal, 6(2), 63–75[10.5121/sipij.2015.6206]
  8. 8.Mohammad Momeny, Ali Mohammad Latif, Mehdi Agha Sarram, Razieh Sheikhpour, Yu Dong Zhang. (2021). A noise robust convolutional neural network for image classification. Results in Engineering, 10, 100225[10.1016/j.rineng.2021.100225]
  9. 9.Kaixuan Wei, Ying Fu, Yinqiang Zheng, Jiaolong Yang. (2021). Physics-based Noise Modeling for Extreme Low-light Photography. IEEE Transactions on Pattern Analysis and Machine Intelligence, 44(11), 1[10.1109/tpami.2021.3103114]
  10. 10.Fan Liu, Qingzeng Song, Guanghao Jin. (2020). The classification and denoising of image noise based on deep neural networks. Applied Intelligence, 50(7), 2194–2207[10.1007/s10489-019-01623-0]
  11. 11.Rusul A. Al Mudhafar, Nidhal K. El Abbadi. (2023). Image Noise Detection and Classification Based on Combination of Deep Wavelet and Machine Learning. Al-Salam Journal for Engineering and Technology, 3(1), 23–36[10.55145/ajest.2024.03.01.003]
  12. 12.F. Maussang, J. Chanussot, A. Hétet, M. Amate. (2007). Higher-Order Statistics for the Detection of Small Objects in a Noisy Background Application on Sonar Imaging. EURASIP Journal on Advances in Signal Processing, 2007(1)[10.1155/2007/47039]
← All publications