AUT Journal of Electrical Engineering

AUT Journal of Electrical Engineering

SaveDoc: A Robust Technique for Degradation Classification on Ancient Document Images

Document Type : Research Article

Authors
Khairun Saddami 1
Rizka Ramadhana 1
Fathurrahman Fathurrahman 2, 1
Khairul Munadi 1
Fitri Arnia 1
1 Department of Electrical and Computer Engineering, Faculty of Engineering, Universitas Syiah Kuala, Banda Aceh, Indonesia
2 Eletrical Engineering Department, College of Engineering and Physics, King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia
10.22060/eej.2026.24823.5773
Abstract
The analysis of degraded ancient documents remains challenging because multiple degradation patterns may appear within a single document image and reduce document quality. In this paper, we propose a combination of convolutional neural networks and extreme learning machine for patch-level degradation classification in ancient document images. The dataset was constructed from 100 source document images and cropped into 700 labeled patches. Four pre-trained architectures, namely VGG19, ResNet101, MobileNetV2, and EfficientNetB0, were used as feature extractors. The number of hidden nodes in the ELM classifier was tuned, and two activation functions, namely radial basis function and sigmoid, were evaluated. The experimental results show that EfficientNetB0 features combined with the extreme learning machine classifier achieved the best performance, with a testing accuracy of 98.21% ± 2.08%. This combination also produced faster inference time compared with the other evaluated models. These results indicate that the EfficientNetB0-extreme learning machine combination is effective for patch-level degradation classification in ancient document images.
Keywords
Degraded Ancient Documents
Degradation Classification
Extreme Learning Machine
Pre-trained CNN features
EfficientNet
Subjects
[1] W. Xiong, X. Jia, J. Xu, Z. Xiong, M. Liu, J. Wang, Historical document image binarization using background estimation and energy minimization, in:  2018 24th International Conference on Pattern Recognition (ICPR), IEEE, 2018, pp. 3716–3721.
[2] Z. Yang, S. Zuo, Y. Zhou, J. He, J. Shi, A review of document binarization: main techniques, new challenges, and trends, Electronics, 13(7) (2024) 1394.
[3] K. Endo, M. Tanaka, M. Okutomi, CNN-based classification of degraded images with awareness of degradation levels, IEEE Transactions on Circuits and Systems for Video Technology, 31(10) (2020) 4046–4057.
[4] A. Shahkolaei, H.Z. Nafchi, S. Al-Maadeed, M. Cheriet, Subjective and objective quality assessment of degraded document images, Journal of Cultural Heritage, 30 (2018) 199–209.
[5] A. Shahkolaei, A. Beghdadi, M. Cheriet, Blind quality assessment metric and degradation classification for degraded document images, Signal Processing: Image Communication, 76 (2019) 11–21.
[6] R.D. Lins, S. Banergee, M. Thielo, Automatically detecting and classifying noises in document images, in: Proceedings of the 2010 ACM Symposium on Applied Computing, 2010, pp. 33–39.
[7] F. Arnia, K. Saddami, K. Munadi, DCNet: Noise-robust convolutional neural networks for degradation classification on ancient documents, Journal of Imaging, 7(7) (2021) 114.
[8] J. Tang, C. Deng, G.-B. Huang, Extreme learning machine for multilayer perceptron, IEEE transactions on Neural Networks and Learning Systems, 27(4) (2015) 809–821.
[9] J. Wang, S. Lu, S.-H. Wang, Y.-D. Zhang, A review on extreme learning machine, Multimedia Tools and Applications, 81(29) (2022) 41611–41660.
[10] G. Huang, G.-B. Huang, S. Song, K. You, Trends in extreme learning machines: A review, Neural Networks, 61 (2015) 32–48.
[11] Y. Qing, Y. Zeng, Y. Li, G.-B. Huang, Deep and wide feature-based extreme learning machine for image classification, Neurocomputing, 412 (2020) 426–436.
[12] R. Kaur, R.K. Roul, S. Batra, A hybrid deep learning CNN-ELM approach for parking space detection in Smart Cities, Neural Computing and Applications, 35(18) (2023) 13665–13683.
[13] M. Jain, W. Andreopoulos, M. Stamp, Convolutional neural networks and extreme learning machines for malware classification, Journal of Computer Virology and Hacking Techniques, 16(3) (2020) 229–244.
[14] Y. Wang, Y. Cao, Z.-J. Zha, J. Zhang, Z. Xiong, Deep degradation prior for low-quality image classification, in:  Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 11049–11058.
[15] K. Saddami, K. Munadi, F. Arnia, Degradation classification on ancient document image based on deep neural networks, in:  2020 3rd International Conference on Information and Communications Technology (ICOIACT), IEEE, 2020, pp. 405–410.
[16] S. Zhou, B. Tan, Electrocardiogram soft computing using hybrid deep learning CNN-ELM, Applied Soft Computing, 86 (2020) 105778.
[17] S. Ali, J. Li, Y. Pei, M.S. Aslam, Z. Shaukat, M. Azeem, An effective and improved CNN-ELM classifier for handwritten digit recognition and classification, Symmetry, 12(10) (2020) 1742.
[18] F. Arnia, K. Saddami, R. Roslidar, R. Muharar, K. Munadi, Towards accurate Diabetic Foot Ulcer image classification: Leveraging CNN pre-trained features and extreme learning machine, Smart Health, 33 (2024) 100502.
[19] L. Alzubaidi, J. Zhang, A.J. Humaidi, A. Al-Dujaili, Y. Duan, O. Al-Shamma, J. Santamaría, M.A. Fadhel, M. Al-Amidie, L. Farhan, Review of deep learning: concepts, CNN architectures, challenges, applications, future directions, Journal of Big Data, 8(1) (2021) 53.
[20] K. Simonyan, A. Zisserman, Very deep convolutional networks for large-scale image recognition, arXiv preprint arXiv:1409.1556,  (2014).
[21] A. Naseer, K. Zafar, Meta‐feature based few‐shot Siamese learning for Urdu optical character recognition, Computational Intelligence, 38(5) (2022) 1707–1727.
[22] M.M. Khayyat, L.A. Elrefaei, M.M. Khayyat, Historical Arabic Images Classification and Retrieval Using Siamese Deep Learning Model, Computers, Materials & Continua, 72(1) (2022).
[23] K. He, X. Zhang, S. Ren, J. Sun, Deep residual learning for image recognition, in:  Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 770–778.
[24] S.A. Siddiqui, A. Dengel, S. Ahmed, Self-supervised representation learning for document image classification, IEEE Access, 9 (2021) 164358–164367.
[25] M. Tang, S. Xie, X. Liu, Ancient character recognition: a novel image dataset of Shui manuscript characters and classification model, Chinese Journal of Electronics, 32(1) (2023) 64–75.
[26] M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, L.-C. Chen, MobileNetV2: Inverted residuals and linear bottlenecks, in:  Proceedings of the IEEE conference on computer vision and pattern recognition, 2018, pp. 4510–4520.
[27] M. Almanea, Deep learning in written Arabic linguistic studies: A comprehensive survey, IEEE Access, 12 (2024) 172196–172233.
[28] Ö. Arslan, S.A. Uymaz, Classification of invoice images by using convolutional neural networks, Journal of Advanced Research in Natural and Applied Sciences, 8(1) (2022) 8–25.
[29] M. Tan, Q. Le, EfficientNet: Rethinking model scaling for convolutional neural networks, in:  International Conference on Machine Learning, PMLR, 2019, pp. 6105–6114.
[30] S. Kanchi, A. Pagani, H. Mokayed, M. Liwicki, D. Stricker, M.Z. Afzal, EmmDocClassifier: Efficient multimodal document image classifier for scarce data, Applied Sciences, 12(3) (2022) 1457.
[31] J. Liu, X. Liu, C. Liu, B.T. Le, D. Xiao, Pruning multilayered ELM using Cholesky factorization and Givens rotation transformation, Mathematical Problems in Engineering, 2021(1) (2021) 5588426.
[32] C. Junyue, D.Q. Zeebaree, C. Qingfeng, D.A. Zebari, Breast cancer diagnosis using hybrid AlexNet-ELM and chimp optimization algorithm evolved by Nelder-Mead simplex approach, Biomedical Signal Processing and Control, 85 (2023) 105053.
[33] B. Gatos, K. Ntirogiannis, I. Pratikakis, ICDAR 2009 document image binarization contest (DIBCO 2009), in:  2009 10th International conference on document analysis and recognition, IEEE, 2009, pp. 1375–1382.
[34] I. Pratikakis, B. Gatos, K. Ntirogiannis, H-DIBCO 2010-handwritten document image binarization competition, in:  2010 12th international conference on frontiers in handwriting recognition, IEEE, 2010, pp. 727–732.
[35] I. Pratikakis, B. Gatos, K. Ntirogiannis, ICDAR 2011 Document Image Binarization Contest (DIBCO 2011), in:  2011 International Conference on Document Analysis and Recognition (ICDAR), IEEE, Beijing, China, 2011, pp. 1506–1510.
[36] I. Pratikakis, B. Gatos, K. Ntirogiannis, ICFHR 2012 competition on handwritten document image binarization (H-DIBCO 2012), in:  2012 international conference on frontiers in handwriting recognition, IEEE, 2012, pp. 817–822.
[37] S.M. Ayatollahi, H.Z. Nafchi, Persian heritage image binarization competition (PHIBC 2012), in:  2013 first Iranian conference on pattern recognition and image analysis (PRIA), IEEE, 2013, pp. 1–4.
[38] I. Pratikakis, B. Gatos, K. Ntirogiannis, ICDAR 2013 document image binarization contest (DIBCO 2013), in:  2013 12th international conference on document analysis and recognition, IEEE, 2013, pp. 1471–1476.
[39] K. Ntirogiannis, B. Gatos, I. Pratikakis, ICFHR2014 competition on handwritten document image binarization (H-DIBCO 2014), in:  2014 14th International conference on frontiers in handwriting recognition, IEEE, 2014, pp. 809–813.
[40] I. Pratikakis, K. Zagoris, G. Barlas, B. Gatos, ICFHR2016 handwritten document image binarization contest (H-DIBCO 2016), in:  2016 15th International Conference on Frontiers in Handwriting Recognition (ICFHR), IEEE, 2016, pp. 619–623.
[41] I. Pratikakis, K. Zagoris, G. Barlas, B. Gatos, ICDAR2017 competition on document image binarization (DIBCO 2017), in:  2017 14Th IAPR International Conference on document analysis and recognition (ICDAR), IEEE, 2017, pp. 1395–1403.
[42] K. Saddami, K. Munadi, S. Muchallil, F. Arnia, Improved thresholding method for enhancing Jawi binarization performance, in:  2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR), IEEE, 2017, pp. 1108–1113.
AUT Journal of Electrical Engineering
Volume 58, Issue 4
2026
Pages 681-694