Analisis pengaruh salt-and-pepper noise terhadap Akurasi Convolutional Neural Network pada Dataset MNIST.
DOI:
https://doi.org/10.25170/jpk.v3i06.8554Keywords:
deep neural network, handwritten digit, robust system designAbstract
This paper discusses the effect of salt-and-pepper noise to the accuracy of deep neural network. We use a standard deep neural network with the convolution as the main component. The original MNIST dataset of 60000 samples of handwritten digit is used. Salt and pepper noise with various ratio was added to create augmented dataset. Training process is set to be stopped when the validation loss exceeds the training loss. As the results, the test accuracy of the system decreases as the noise level increases. While in the train process, this trend does not occur
References
Bachri, K. O. (2025). Performance Analysis of Deep Neural Network to Noisy Digit Dataset. Jurnal Elektro, 18(1), 28–40.
Cai, R. T., Chen, G. X., Li, J., Du, R. S., Lu, H. Y., Qi, Y. L., & Chen, J. X. (2025). ADenoising Method for Salt and Pepper Noise in Remote Sensing Based on Swin-Transformer Convolution U-Net and Filtering—FSCU-Net. Earth and Space Science, 12(8),1-27. https://doi.org/10.1029/2025EA004225
Egmont-Petersen, M., Talmon, J. L., & Hasman, A. (1997). Robustness metrics for measuring the influence of additive noise on the performance of statistical classifiers. International Journal of Medical Informatics, 46(2), 103–112. https://doi.org/https://doi.org/10.1016/S1386-5056(97)00062-2
Horé, A., & Ziou, D. (2010). Image Quality Metrics: PSNR vs. SSIM. 2010 20th International Conference on Pattern Recognition, (pp.2366–2369). https://doi.org/10.1109/ICPR.2010.579
Ilesanmi, A. E., & Ilesanmi, T. O. (2021). Methods for image denoising using convolutional neural network: a review. Complex and Intelligent Systems, 7(5), 2179–2198. https://doi.org/10.1007/s40747-021-00428-4
Islam, M. J. (2025). Adaptive weight-modified Riesz mean filter for high-density salt-and-pepper noise removal. Journal of Electrical Systems and Information Technology, 12(1),1-16. https://doi.org/10.1186/s43067-025-00266-1
Kamran, S. A., Moghnieh, H., Hossain, K. F., Bartlett, A., Tavakkoli, A., Drumm, B. T., Sanders, K. M., & Baker, S. A. (2024). Automated denoising software for calcium imaging signals using deep learning. Heliyon, 10(21) e39574. https://doi.org/10.1016/j.heliyon.2024.e39574
Kavitha, G., Prakash, C., Alhomrani, M., Pradeep, N., Alamri, A. S., Pareek, P. K., & Alhassan, M. (2023). Noise Estimation and Type Identification in Natural Scene and Medical Images using Deep Learning Approaches. Contrast Media & Molecular Imaging, 2023, 1–15. https://doi.org/10.1155/2023/3923667
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998). Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11), 2278–2324. https://doi.org/10.1109/5.726791
Li, B., Zhang, Y., Chen, W., Ma, Y., & Li, L. (2024). Neural Network-Based Investigation of Periodic Noise Reduction Methods for High-Resolution Infrared Line Scanning Images. Remote Sensing, 16(5), 1-24. https://doi.org/10.3390/rs16050841
Li, G., Zrimec, J., Ji, B., Geng, J., Larsbrink, J., Zelezniak, A., Nielsen, J., & Engqvist, M. K. M. (2021). Performance of Regression Models as a Function of Experiment Noise. Bioinformatics and Biology Insights, 15. https://doi.org/10.1177/11779322211020315
Li, R., Liu, T., & Yin, J. (2024). An encryption algorithm for color images based on an improved dual-chaotic system combined with DNA encoding. Scientific Reports, 14(1). https://doi.org/10.1038/s41598-024-71267-9
Lin, J., Qin, J., & Lu, S. (2022). Suppressing Shot Noise Using Quadratic Variable Step-Size Quantization for the Initial Acquisition of Camera-Raw Image Data. IEEE Transactions on Image Processing, 31, 5242–5256. https://doi.org/10.1109/TIP.2022.3184256
Nazir, N., Sarwar, A., & Saini, B. S. (2024). Recent developments in denoising medical images using deep learning: An overview of models, techniques, and challenges. Micron, 180, 103615. https://doi.org/https://doi.org/10.1016/j.micron.2024.103615
Rahman, S., Pal, S., Yearwood, J., & Karmakar, C. (2024). Robustness of Deep Learning models in electrocardiogram noise detection and classification. Computer Methods and Programs in Biomedicine, 253. https://doi.org/10.1016/j.cmpb.2024.108249
Rani, A., & Bhogal, R. K. (2025). Efficient real-world image denoising using multi-scale gaussian pyramids. Scientific Reports, 15(1). https://doi.org/10.1038/s41598-025-23942-8
Sáez, J. A., Luengo, J., & Herrera, F. (2016). Evaluating the classifier behavior with noisy data considering performance and robustness: The Equalized Loss of Accuracy measure. Neurocomputing, 176, 26–35. https://doi.org/https://doi.org/10.1016/j.neucom.2014.11.086
Senthamizh Selvi, R., Aarthi Varshini, B., & Deekshiga, S. (2023). A Comparative Analysis of Image Denoising Filters for Salt and Pepper Noise. 2023 9th International Conference on Smart Structures and Systems (ICSSS), (pp.1–10). https://doi.org/10.1109/ICSSS58085.2023.10407368
Shi, B., Cao, T., Wang, Z., Ge, Q., & Guo, W. (2023). Research on Sonar Image Denoising Based on Optimized Spatial Pixel Ranking Algorithm. 2023 8th International Conference on Signal and Image Processing (ICSIP), (pp. 249–253). https://doi.org/10.1109/ICSIP57908.2023.10271087
Wu, S., Guo, C., & Wang, X. (2023). Salt-and-pepper Noise Reduction for Medical Images based on Image Fusion. Current Medical Imaging Formerly Current Medical Imaging Reviews, 20,1-13. https://doi.org/10.2174/1573405620666230525104841
Xing, Y., Xu, J., Tan, J., Li, D., & Zha, W. (2019). Deep CNN for removal of salt and pepper noise. IET Image Processing, 13(9), 1550–1560. https://doi.org/10.1049/iet-ipr.2018.6004
