Hyperparameter Analysis of Adam-Optimized Deep Transfer Learning for Indonesian Banknote-Denomination Recognition
Abstract
Accurately distinguishing Indonesian banknote denominations remains challenging for older adults and people with visual impairments, especially after the recent release of a redesigned currency. To address this gap, we propose an intelligent recognition system based on deep transfer learning with systematic hyperparameter optimization. Our approach fine-tunes a pre-trained ResNet-50 backbone while simultaneously calibrating the learning rate, batch size, and training epochs via an am-optimized grid search. Extensive experiments across multiple convolutional neural-network architectures confirmed that the ResNet-50 model, trained with a learning rate of 0.0001, batch sizes 20 and 15 epochs, achieved state-of-the-art performance: 99.29 % accuracy, 99.32 % precision, 99.29 % recall, and 99.29 % F1-score. These findings demonstrate that carefully tuned transfer-learning pipelines can deliver near-perfect classification of the new Indonesian banknote series while retaining strong generalization to previously unseen images, thereby offering a practical assistive tool for the visually impaired and elderly.
Keywords: Deep transfer learning; ResNet-50; Convolutional neural networks; Indonesian banknote recognition; Hyper-parameter optimization; Assistive computer vision.
Full Text:
PDFReferences
M. I. Lubis, “Online buying and selling transactions under international private law,” J. Priv. Commer. Law, vol. 2, no. 1, 2018, doi: 10.15294/jpcl.v2i1.14499. https://doi.org/10.15294/jpcl.v2i1.14499
Y. Chen, S. Wang, L. Lin, Z. Cui, and Y. Zong, “Computer Vision and Deep Learning Transforming Image Recognition and Beyond,” Int. J. Comput. Sci. Inf. Technol., vol. 2, no. 1, 2024, doi: 10.62051/ijcsit.v2n1.06. https://doi.org/10.62051/ijcsit.v2n1.06
J. Chai, H. Zeng, A. Li, and E. W. T. Ngai, “Deep learning in computer vision: A critical review of emerging techniques and application scenarios,” Mach. Learn. with Appl., vol. 6, 2021, doi: 10.1016/j.mlwa.2021.100134. https://doi.org/10.1016/j.mlwa.2021.100134
Golani Hema Pribhdas, “Deep Learning for Convolution Neural Networks and Recurrent Neural Networks: A Review,” Recent trends Manag. Commer., vol. 2, no. 4, 2021, doi: 10.46632/rmc/2/4/2. https://doi.org/10.46632/rmc/2/4/2
J. M. Górriz et al., “Computational approaches to Explainable Artificial Intelligence: Advances in theory, applications and trends,” Inf. Fusion, vol. 100, p. 101945, Dec. 2023, doi: 10.1016/J.INFFUS.2023.101945. https://doi.org/10.1016/j.inffus.2023.101945
S. F. Ahmed et al., “Deep learning modelling techniques: current progress, applications, advantages, and challenges,” Artif. Intell. Rev., vol. 56, no. 11, 2023, doi: 10.1007/s10462-023-10466-8. https://doi.org/10.1007/s10462-023-10466-8
U. Sadyk, C. Turan, and R. Baimukashev, “Overview of Deep Learning Models for Banknote Recognition,” 2023, doi: 10.1109/ICECCO58239.2023.10147142. https://doi.org/10.1109/icecco58239.2023.10147142
T. D. Pham, D. T. Nguyen, W. Kim, S. H. Park, and K. R. Park, “Deep learning-based banknote fitness classification using the reflection images by a visible-light one-dimensional line image sensor,” Sensors (Switzerland), vol. 18, no. 2, 2018, doi: 10.3390/s18020472. https://doi.org/10.3390/s18020472
L. Alzubaidi et al., “Review of deep learning: concepts, CNN architectures, challenges, applications, future directions,” J. Big Data 2021 81, vol. 8, no. 1, pp. 1–74, Mar. 2021, doi: 10.1186/S40537-021-00444-8. https://doi.org/10.1186/s40537-021-00444-8
R. Ramadhan, J. Y. Sari, and I. P. Ningrum, “Identification of Authenticity and Nominal Value of Indonesia Banknotes Using Fuzzy KNearest Neighbor Method,” IJNMT (International J. New Media Technol., vol. 6, no. 1, pp. 32–37, Aug. 2019, doi: 10.31937/ijnmt.v6i1.989. https://doi.org/10.31937/ijnmt.v6i1.989
T. D. Pham, C. Park, D. T. Nguyen, G. Batchuluun, and K. R. Park, “Deep Learning-Based Fake-Banknote Detection for the Visually Impaired People Using Visible-Light Images Captured by Smartphone Cameras,” IEEE Access, vol. 8, 2020, doi: 10.1109/ACCESS.2020.2984019. https://doi.org/10.1109/access.2020.2984019
I. M. S. Kumara, G. P. R. S. Jati, and N. P. W. Yuniari, “Integrate Yolov8 Algorithm For Rupiah Denomination Detection In All-In-One Smart Cane For Visually Impaired,” Techno.Com, vol. 23, no. 1, pp. 176–186, Feb. 2024, doi: 10.62411/tc.v23i1.9734. https://doi.org/10.62411/tc.v23i1.9734
N. Mallesh et al., “Knowledge transfer to enhance the performance of deep learning models for automated classification of B cell neoplasms,” Patterns, vol. 2, no. 10, 2021, doi: 10.1016/j.patter.2021.100351. https://doi.org/10.1016/j.patter.2021.100351
V. Meshram, K. Patil, and V. Meshram, “Evaluation of Top Pretrained Models Using Transfer Learning on Banknote Dataset with Quality Parameter,” Ing. des Syst. d’Information, vol. 28, no. 3, 2023, doi: 10.18280/isi.280319. https://doi.org/10.18280/isi.280319
A. H. M. Linkon, M. M. Labib, F. H. Bappy, S. Sarker, M. E. Jannat, and M. S. Islam, “Deep learning approach combining lightweight CNN architecture with transfer learning: An automatic approach for the detection and recognition of bangladeshi banknotes,” 2020, doi: 10.1109/ICECE51571.2020.9393113. https://doi.org/10.18280/isi.280319
A. Yildiz, A. A. Almisreb, Š. Dzakmic, and N. M. Tahir, “Banknotes Counterfeit Detection Using Deep Transfer Learning Approach,” Int. J. Adv. Trends Comput. Sci. Eng., vol. 9, no. 5, 2020, doi: 10.30534/ijatcse/2020/172952020. https://doi.org/10.30534/ijatcse/2020/172952020
A. Khalil et al., “Mobile Deep Classification of UAE Banknotes for the Visually Challenged,” 2022, doi: 10.1109/FiCloud57274.2022.00053. https://doi.org/10.1109/ficloud57274.2022.00053
M. B. Alejo, J. L. D. Villanueva, M. P. E. Garchitorena, S. C. Reyes, J. M. B. Delos Reyes, and Q. A. L. Marasigan, “Philippine Banknote Counterfeit Detection through Domain Adaptive Deep Learning Model of the Convolutional Neural Network,” Int. J. Comput. Digit. Syst., vol. 13, no. 1, 2023, doi: 10.12785/ijcds/130103. https://doi.org/10.12785/ijcds/130103
T. Wagner and S. Sommer, “Feature Based Bearing Fault Detection With Phase Current Sensor Signals Under Different Operating Conditions,” PHM Soc. Eur. Conf., vol. 6, no. 1, 2021, doi: 10.36001/phme.2021.v6i1.2852. https://doi.org/10.36001/phme.2021.v6i1.2852
K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2016, vol. 2016-Decem, doi: 10.1109/CVPR.2016.90. https://doi.org/10.1109/cvpr.2016.90
E. Rendón, R. Alejo, C. Castorena, F. J. Isidro-Ortega, and E. E. Granda-Gutiérrez, “Data sampling methods to dealwith the big data multi-class imbalance problem,” Appl. Sci., vol. 10, no. 4, 2020, doi: 10.3390/app10041276. https://doi.org/10.3390/app10041276
A. K. Azlim Khan and N. H. Ahamed Hassain Malim, “Comparative Studies on Resampling Techniques in Machine Learning and Deep Learning Models for Drug-Target Interaction Prediction,” Molecules, vol. 28, no. 4. 2023, doi: 10.3390/molecules28041663. https://doi.org/10.3390/molecules28041663
S. Li, Q. Li, and M. Li, “A Method for Network Intrusion Detection Based on GAN-CNN-BiLSTM,” Int. J. Adv. Comput. Sci. Appl., vol. 14, no. 5, 2023, doi: 10.14569/IJACSA.2023.0140554. https://doi.org/10.14569/ijacsa.2023.0140554
T. Sulistyowati, P. Purwanto, F. Alzami, and R. A. Pramunendar, “VGG16 Deep Learning Architecture Using Imbalance Data Methods For The Detection Of Apple Leaf Diseases,” Monet. J. Keuang. dan Perbank., vol. 11, no. 1, 2023, doi: 10.32832/moneter.v11i1.57. https://doi.org/10.32832/moneter.v11i1.57
T. M. Alam et al., “An Efficient Deep Learning-Based Skin Cancer Classifier for an Imbalanced Dataset,” Diagnostics, vol. 12, no. 9, 2022, doi: 10.3390/diagnostics12092115. https://doi.org/10.3390/diagnostics12092115
M. O. Ojo and A. Zahid, “Improving Deep Learning Classifiers Performance via Preprocessing and Class Imbalance Approaches in a Plant Disease Detection Pipeline,” Agronomy, vol. 13, no. 3, 2023, doi: 10.3390/agronomy13030887. https://doi.org/10.3390/agronomy13030887
A. Y. Hussein, P. Falcarin, and A. T. Sadiq, “Enhancement performance of random forest algorithm via one hot encoding for IoT IDS,” Period. Eng. Nat. Sci., vol. 9, no. 3, 2021, doi: 10.21533/pen.v9i3.2204. https://doi.org/10.21533/pen.v9i3.2204
B. Gu and Y. Sung, “Enhanced reinforcement learning method combining one-hot encoding-based vectors for cnn-based alternative high-level decisions,” Appl. Sci., vol. 11, no. 3, 2021, doi: 10.3390/app11031291. https://doi.org/10.3390/app11031291
S. Bagui, D. Nandi, S. Bagui, and R. J. White, “Machine Learning and Deep Learning for Phishing Email Classification using One-Hot Encoding,” J. Comput. Sci., vol. 17, no. 7, 2021, doi: 10.3844/jcssp.2021.610.623. https://doi.org/10.3844/jcssp.2021.610.623
M. Susanty, R. Hertadi, A. Purwarianti, and T. L. E. Rajab, “Low Complexity Classification of Thermophilic Protein using One Hot Encoding as Protein Representation,” Int. J. Adv. Comput. Sci. Appl., vol. 13, no. 12, 2022, doi: 10.14569/IJACSA.2022.0131212. https://doi.org/10.14569/ijacsa.2022.0131212
L. Yu, R. Zhou, R. Chen, and K. K. Lai, “Missing Data Preprocessing in Credit Classification: One-Hot Encoding or Imputation?,” Emerg. Mark. Financ. Trade, vol. 58, no. 2, 2022, doi: 10.1080/1540496X.2020.1825935. https://doi.org/10.1080/1540496x.2020.1825935
T. S. Prajwal and A. K. Ilavarasi, “A Comparative Study of RESNET-Pretrained Models for Computer Vision,” ACM Int. Conf. Proceeding Ser., pp. 419–425, Aug. 2023, doi: 10.1145/3607947.3608042. https://doi.org/10.1145/3607947.3608042
L. N. Smith and N. Topin, “Super-convergence: very fast training of neural networks using large learning rates,” 2019, doi: 10.1117/12.2520589. https://doi.org/10.1117/12.2520589
J. Heaton, “Ian Goodfellow, Yoshua Bengio, and Aaron Courville: Deep learning The MIT Press, 2016, 800 pp, ISBN: 0262035618,” Genet. Program. Evolvable Mach., vol. 19, no. 1–2, 2018.
Refbacks
- There are currently no refbacks.
