ADAPTIVE CLASS WEIGHTING DAN AUGMENTATION UNTUK KLASIFIKASI BATIK KERATON
DOI:
https://doi.org/10.51903/informatika.v6i1.1516Keywords:
Batik Keraton, imbalanced dataset, adaptive class weighting, data augmentation, transfer learning, ResNet50V2.Abstract
This study aims to improve the performance of Batik Keraton motif classification on an imbalanced dataset through the integration of adaptive class weighting and data augmentation within a transfer learning framework. The dataset consists of 1,799 images across four classes (Kawung, Mega
Mendung, Parang, Truntum), preprocessed to 224×224 pixels and split stratifiedly into training, validation, and test sets (80/10/10). Three transfer learning architectures—ResNet50V2, VGG16, and EfficientNetB0—were evaluated with adaptive class weighting and geometric augmentation to enhance minority-class representation. The results indicate that ResNet50V2 with pretrained weights achieved the best performance, reaching a test accuracy of 92.78%, macro precision of 93.13%, macro recall of 92.79%, and a macro F1-score of 92.83%. Adaptive class weighting improved sensitivity toward minority classes, while augmentation contributed to model stability and generalization. These findings demonstrate that combining adaptive weighting and augmentation effectively enhances Batik Keraton motif classification under imbalanced data conditions.
References
DAFTAR PUSTAKA
[1] E. Winarno, A. Solichan, A. P. Ramdani, W. Hadikurniawati, and A. Septiarini, “Enhanced Semarang batik classification using deep learning: a comparative study of CNN architectures,” Bulletin of Electrical Engineering and Informatics, vol. 14, no. 5, p. 9347, 2023, doi: 10.11591/eei.v14i5.9347.
[2] D. Sinaga, C. Jatmoko, Suprayogi, and N. Hendriyanto, “Multi‐Layer Convolutional Neural Networks for Batik Image Classification,” Scientific Journal of Informatics, vol. 11, no. 2, pp. 477–484, 2024, doi: 10.15294/sji.v11i2.3309.
[3] L. Fitriani, D. Tresnawati, and M. B. Sukriyansah, “Image Classification on Garutan Batik Using Convolutional Neural Network with Data Augmentation,” JUITA: Jurnal Informatika, vol. 11, no. 1, pp. 107–115, 2023, doi: 10.30595/juita.v11i1.16166.
[4] N. Liu, J. Wang, Y. Zhu, L. Wan, and Q. Li, “Improving imbalance classification via ensemble learning based on two-stage learning,” Front Comput Neurosci, vol. 17, p. 1296897, 2024, doi: 10.3389/fncom.2023.1296897.
[5] M.-A. Lavoie and S. Waslander, “Class instance balanced learning for long-tailed classification,” 2023. doi: 10.48550/arXiv.2307.05322.
[6] A. Galdran, G. Carneiro, and M. A. González Ballester, “Balanced-MixUp for highly imbalanced medical image classification,” 2021. doi: 10.48550/arXiv.2109.09850.
[7] F. Wulandari, A. Rahmawati, and L. E. Nugroho, “Enhancing Batik image classification through data augmentation and transfer learning,” Journal of King Saud University – Computer and Information Sciences, vol. 35, no. 8, p. 102547, 2023, doi: 10.1016/j.jksuci.2023.102547.
[8] M. Buda, A. Maki, and M. A. Mazurowski, “A systematic study of the class imbalance problem in convolutional neural networks,” Neural Networks, vol. 143, pp. 140–150, 2021, doi: 10.1016/j.neunet.2021.06.014.
[9] A. Setiawan, E. Prasetyo, and R. Hidayat, “Automated classification of Indonesian Batik motifs using EfficientNet and MobileNetV2 architectures,” IEEE Access, vol. 10, pp. 78901–78912, 2022, doi: 10.1109/ACCESS.2022.3189012.
[10] S. Huang, Z. Li, and H. Zhang, “Transfer learning-based textile defect classification using convolutional neural networks,” Comput Ind, vol. 132, p. 103516, 2021, doi: 10.1016/j.compind.2021.103516.
[11] D. Dablain, K. N. Jacobson, C. Bellinger, M. Roberts, and N. V Chawla, “Understanding CNN fragility when learning with imbalanced data,” Mach Learn, vol. 113, pp. 4785–4810, 2024, doi: 10.1007/s10994-023-06326-9.
[12] W. Chen, K. Yang, Z. Yu, Y. Shi, and C. L. P. Chen, “A survey on imbalanced learning: latest research, applications and future directions,” Artif Intell Rev, vol. 57, p. 137, 2024, doi: 10.1007/s10462-024-10759-6.
[13] M. Carvalho, A. J. Pinho, and S. Brás, “Resampling approaches to handle class imbalance: a review from a data perspective,” J Big Data, vol. 12, p. Article 71, 2025, doi: 10.1186/s40537-025-01119-4.
[14] C. Zhang, H. Li, and et al., “Multimodal framework for long-tailed recognition,” Applied Sciences, vol. 14, no. 22, p. 10572, 2024, doi: 10.3390/app142210572.
[15] Y. Xiang and et al., “Enhancing robustness in long-tailed image classification with angular approaches and balanced learning,” Data Sci Eng, vol. 10, pp. 480–494, 2025, doi: 10.1007/s41019-025-00286-x.
[16] E. Sala and et al., “A data augmentation methodology to reduce the class imbalance in histopathology images,” Journal of Imaging Informatics in Medicine, vol. 37, pp. 1767–1782, 2024, doi: 10.1007/s10278-024-01018-9.
[17] M. Dorozynski and F. Rottensteiner, “Deep descriptor learning with auxiliary classification loss for retrieving images of silk fabrics in the context of preserving European silk heritage,” ISPRS Int J Geoinf, vol. 11, no. 2, p. 82, 2022, doi: 10.3390/ijgi11020082.
[18] C. Dionysiou, P. Davnos, and et al., “From contemporary datasets to cultural heritage performance: Explainability and energy profiling of visual models towards textile identification,” Heritage, vol. 8, no. 11, p. 447, 2024.
[19] I. Radosavovic, J. Johnson, S. Xie, and P. Dollár, “Design of Experiments for Deep Learning: Benchmarking and Analysis,” IEEE Trans Pattern Anal Mach Intell, vol. 45, no. 2, pp. 1600–1615, 2022, doi: 10.1109/TPAMI.2022.3178449.
[20] A. Pala, A. Oleynik, I. Utseth, and N. O. Handegard, “Addressing class imbalance in deep learning for acoustic target classification,” ICES Journal of Marine Science, 2023, doi: https://doi.org/10.1093/icesjms/fsad165.
[21] R. Walsh and M. Tardy, “A comparison of techniques for class imbalance in deep learning classification of breast cancer,” Diagnostics, vol. 13, no. 1, p. 67, 2022, doi: https://doi.org/10.3390/diagnostics13010067.
[22] E. Çınar, “Comparative analysis of deep learning-based feature extraction and traditional classification approaches for tomato disease detection,” Agronomy, vol. 15, no. 7, p. 1509, 2024, doi: https://doi.org/10.3390/agronomy15071509.
[23] D. Yuliani, A. Widodo, and A. Hidayanto, “Batik motif classification using transfer learning with ResNet50 and VGG16 models,” Applied Sciences, vol. 13, no. 4, p. 2761, 2023, doi: 10.3390/app13042761.
[24] A. Tharwat, “Classification assessment methods applied on machine learning algorithms for medical diagnosis,” Comput Biol Med, vol. 132, p. 104287, 2021, doi: 10.1016/j.compbiomed.2021.104287.
[25] T. H. Nguyen, D. H. Pham, and Q. H. Tran, “Deep transfer learning for textile fabric pattern classification based on multi-feature fusion,” J Imaging, vol. 10, no. 3, p. 45, 2024, doi: 10.3390/jimaging10030045.
[26] J. M. Johnson and T. M. Khoshgoftaar, “Survey on deep learning with class imbalance,” J Big Data, vol. 9, no. 1, p. 3, 2022, doi: 10.1186/s40537-021-00537-3.
[27] S. Wang, H. Xu, H. Liu, and X. Wang, “Adaptive loss reweighting and data augmentation for imbalanced image classification,” IEEE Access, vol. 11, pp. 49213–49225, 2023, doi: 10.1109/ACCESS.2023.3268129.
[28] Y. Cui, M. Jia, T.-Y. Lin, Y. Song, and S. Belongie, “Class-balanced loss based on effective number of samples,” IEEE Trans Pattern Anal Mach Intell, vol. 43, no. 6, pp. 2199–2212, 2021, doi: 10.1109/TPAMI.2020.3018749.









.png)