MULTI-OBJECTIVE OPTIMIZATION OF RESNET50 ARCHITECTURE USING GENETIC ALGORITHM FOR ENHANCED BATIK MOTIF CLASSIFICATION
DOI:
https://doi.org/10.33480/jitk.v12i1.8050Keywords:
Batik Classification, Genetic Algorithm, Multi-Objective Optimization, ResNet50, Transfer LearningAbstract
Automatic batik motif classification remains challenging due to the high visual similarity among many motifs, making manual differentiation difficult. This study proposes a weighted-sum multi-objective Genetic Algorithm (GA) to optimize a pre-trained ResNet50 model for batik motif classification. The novelty of this study lies in optimizing the top-layer architecture and transfer-learning hyperparameters of ResNet50 by jointly considering classification accuracy, trainable parameter count, and inference latency. Unlike many previous GA-based CNN studies that mainly focus on accuracy or construct architectures from scratch, this study employs GA as a resource-efficient optimizer for practical deployment. The research used a quantitative experimental design with a secondary dataset of 3,550 batik images from five motif classes, namely Kawung, Megamendung, Parang, Sidomukti, and Truntum. The optimization process searched for the best configuration of four hyperparameters, namely the number of neurons in the dense layer, dropout rate, learning rate, and fine-tuning depth, while Adam was used as a fixed optimizer throughout the experiments. The results show that the proposed model improved classification accuracy from 84.00% to 90.00%, reduced trainable parameters by 94.4% from 23.59 million to 1.31 million, and decreased inference time by 29.4% compared with the baseline ResNet50 model. These findings indicate that the proposed method can achieve a favorable balance between predictive performance and computational efficiency for cultural heritage recognition on resource-constrained devices.
Downloads
References
[1] M. Syafitri, “Recognizing the Batik Heritage of the Archipelago : From Coastal to Inland,” vol. 1, no. January, 2025.
[2] W. Wesnina, M. Prabawati, and M. Noerharyono, “Integrating traditional and contemporary in digital techniques : the analysis of Indonesian batik motifs evolution,” Cogent Arts Humanit., vol. 12, no. 1, p., 2025, doi: 10.1080/23311983.2025.2474845.
[3] H. Sastypratiwi and H. Muhardi, “Batik Recognition and Classification Using Transfer Learning and MobileNet Approach,” vol. 8, no. December, 2024.
[4] A. Jordan et al., “Improving Batik Classification using CNN Advanced Improving Batik Pattern Pattern Classification on using CNN with Advanced Augmentation and Oversampling Imbalanced Dataset Improving Batik Pattern Classification using CNN with Advanced Augmentation a,” Procedia Comput. Sci., vol. 227, pp. 508–517, 2023, doi: 10.1016/j.procs.2023.10.552.
[5] D. Gede, T. Meranggi, N. Yudistira, and Y. A. Sari, “International Journal On Informatics Visualization Journal Homepage : Www.Joiv.Org/Index.Php/Joiv International Journal On Informatics Visualization Batik Classification Using Convolutional Neural Network with Data Improvements,” vol. 6, no. March, pp. 6–11, 2022, [Online]. Available: www.joiv.org/index.php/joiv
[6] G. Rangel, J. C. Cuevas-tello, A. G. Silva-trujillo, J. Nunez-varela, and C. Puente, “Review Article A Survey on Convolutional Neural Networks and Their Performance Limitations in Image Recognition Tasks,” vol. 2024, 2024, doi: 10.1155/2024/2797320.
[7] T. Shi, Z. Hu, J. Yu, and K. Jiang, “Improved Method of ResNet50 Image Classification Based on Transfer Learning,” vol. 10, no. 02, 2025, doi: 10.2478/ijanmc-2025-0011.
[8] V. Y. Cambay et al., “Automated Detection of Gastrointestinal Diseases Using Resnet50 * -Based Explainable Deep Feature Engineering Model with Endoscopy Images,” 2024.
[9] L. Ferreira, R. André, R. Backes, B. Augusto, and N. Travençolo, “Optimizing a Deep Residual Neural Network with Genetic Algorithm for Acute Lymphoblastic Leukemia Classification,” J. Digit. Imaging, pp. 623–637, 2022, doi: 10.1007/s10278-022-00600-3.
[10] K. Lehana, “A Genetic Algorithm Based Optimized Convolutional Neural Network For Face Recognition,” vol. 33, no. 1, pp. 21–31, 2023, doi: 10.34768/amcs-2023-0002.
[11] T. Goel, R. Murugan, S. Mirjalili, and D. K. Chakrabartty, “Multi-COVID-Net: Multi-objective optimized network for COVID-19 diagnosis from chest X-ray images,” Appl. Soft Comput., vol. 115, p. 108250, 2022, doi: 10.1016/j.asoc.2021.108250.
[12] N. T. Sarshar, S. Sadeghi, M. Kamsari, M. Avazpour, S. J. Ghoushchi, and R. Ranjbarzadeh, “Advancing Brain MRI Image Classification : Integrating VGG16 and ResNet50 with a Multi-Verse Optimization Method,” no. Ml, pp. 499–523, 2024.
[13] X. Wang, Q. Ye, L. Liu, H. Niu, and B. Du, “ResNet-50-NTS digital painting image style classification based on Three-Branch convolutional attention,” Egypt. Informatics J., vol. 29, no. November 2024, p. 100614, 2025, doi: 10.1016/j.eij.2025.100614.
[14] K. Irfan, “Performance Analysis of ResNet50 on Balanced and Imbalanced Image Classification Datasets,” 2022.
[15] M. Amine, D. Lamrani, S. Saleh, and W. Moutaouakil, “Intelligence-Based Medicine Optimizing ResNet50 performance using stochastic gradient descent on MRI images for Alzheimer ’ s disease classification,” vol. 11, no. October 2024, 2025.
[16] A. Du, Q. Zhou, and Y. Dai, “applied sciences Methodology for Evaluating the Generalization of ResNet,” pp. 1–18, 2024.
[17] R. Zhang, Y. Zhu, and Z. Ge, “Transfer Learning for Leaf Small Dataset Using Improved ResNet50 Network with Mixed Activation Functions,” 2022.
[18] M. M. Ansari et al., “Evaluating CNN Architectures and Hyperparameter Tuning for Enhanced Lung Cancer Detection Using Transfer Learning,” vol. 2024, 2024, doi: 10.1155/2024/3790617.
[19] Y. An, C. Zhang, J. Shao, Y. Yan, and B. Sun, “An Efficient Evolutionary Neural Architecture Search Algorithm Without Training,” pp. 1–26, 2025.
[20] F. Y. Santoso, E. Sediyono, and H. D. Purnomo, “Optimalisasi Hyper Parameter Convolutional Neural Networks Hyperparameter Optimization Using Ant Colony Optimization,” Vol. 11, No. 2, Pp. 243–248, 2024, doi: 10.25126/jtiik.20241127105.
[21] M. Munsarif, E. Noersasongko, and P. Nurtantio, “Improving convolutional neural network based on hyperparameter optimization using variable length genetic algorithm for english digit handwritten recognition,” vol. 9, no. 1, pp. 66–78, 2023.
[22] C. Ding, S. Zhou, N. Liu, and B. Yuan, “Resource-Aware Neural Architecture Search for Efficient Deep Learning on Edge Devices,” IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 42, no. 8, pp. 2510–2523, 2023, doi: 10.1109/TCAD.2022.3228891.
[23] S. M. Shariatzadeh, M. Fathy, R. Berangi, and M. Shahverdy, “A Survey on Multi-Objective Neural Architecture Search,” arXiv preprint arXiv:2307.09099, pp. 1–22, 2023, doi: 10.48550/arXiv.2307.09099.
[24] A. Azedou, A. Amine, I. Kisekka, and S. Lahssini, “Genetic algorithm optimization of ensemble learning approach for improved land cover and land use mapping : Application to Talassemtane National Park,” Ecol. Indic., vol. 177, no. March 2024, p. 113776, 2025, doi: 10.1016/j.ecolind.2025.113776.
[25] V. Kralev, “Combining Genetic Algorithm with Local Search Method in Solving Optimization Problems,” 2024.
[26] Y. Zhao, “Multi-Objective Neural Architecture Search by Learning Search Space Partitions,” vol. 25, pp. 1–41, 2024.
[27] J. Darling, P. Rosita, and W. S. Jacob, “International Journal of Intelligent Networks Multi-Objective Genetic Algorithm and CNN-Based Deep Learning Architectural Scheme for effective spam detection,” Int. J. Intell. Networks, vol. 3, no. February, pp. 9–15, 2022, doi: 10.1016/j.ijin.2022.01.001.
[28] S. Siddiqui, C. Kyrkou, and T. Theocharides, “Efficient Global Neural Architecture Search,” SN Comput. Sci., vol. 6, no. 3, 2025, doi: 10.1007/s42979-025-03790-z.
[29] S. Salmani, P. Avval, N. D. Eskue, R. M. Groves, and V. Yaghoubi, “Systematic review on neural architecture search,” no. January, 2025.
[30] G. Patrucco, F. Setragno, and A. Spanò, “Synthetic Training Datasets for Architectural Conservation: A Deep Learning Approach for Decay Detection,” Remote Sens., vol. 17, no. 10, 2025, doi: 10.3390/rs17101714.
[31] R. F. Alya, “Classification Of Batik Motif Using Transfer Learning On Convolutional Neural Network (Cnn),” J. Tek. Inform., Vol. 4, No. 1, pp. 161–170.
[32] M. Utami, E. Ermatita, and A. Abdiansah, “Improving the transfer learning for batik besurek textile motif classification,” vol. 14, no. 4, pp. 3172–3181, 2025, doi: 10.11591/ijai.v14.i4.pp3172-3181.
[33] C. Li et al., “Genetic algorithm based hyper-parameters optimization for transfer convolutional neural network,” p. 46, 2022, doi: 10.1117/12.2637170.
[34] N. Sinha, A. El Rahman Shabayek, A. Kacem, P. Rostami, C. Shneider, and D. Aouada, “Hardware Aware Evolutionary Neural Architecture Search using Representation Similarity Metric,” Proc. - 2024 IEEE Winter Conf. Appl. Comput. Vision, WACV 2024, pp. 2616–2625, 2024, doi: 10.1109/WACV57701.2024.00261.
[35] F. A. Prasetyo, A. F. Hidayatullah, and I. Fitri, “Indonesian Batik Classification Using Convolutional Neural Network with Transfer Learning,” Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi), vol. 7, no. 2, pp. 312–318, 2023, doi: 10.29207/resti.v7i2.4821.
[36] Dionisiusdh, “Indonesian Batik Motifs Dataset,” Kaggle, 2023. [Online]. Available: https://www.kaggle.com/datasets/dionisiusdh/indonesian-batik-motifs. [Accessed: 15-Jan-2026].
Downloads
Published
Issue
Section
License
Copyright (c) 2026 P.A.M. ZIDANE R.W.P.P ZER ZER, Dedy Hartama, Agus Perdana Windarto
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
-a.jpg)
-b.jpg)
