FINE-TUNING RESNET50V2 WITH ADAMW AND ADAPTIVE TRANSFER LEARNING FOR SONGKET CLASSIFICATION IN LOMBOK

Erfan  Wahyudi; Bahtiar Imran; Zaeniah; Surni  Erniwati; Muh Nasirudin  Karim; Zumratul Muahidin

doi:10.33480/pilar.v21i1.6485

Authors

Erfan Wahyudi Institut Pemerintahan Dalam Negeri
Bahtiar Imran Universitas Teknologi Mataram
Zaeniah Universitas Teknologi Mataram
Surni Erniwati Universitas Teknologi Mataram
Muh Nasirudin Karim Universitas Teknologi Mataram
Zumratul Muahidin Universitas Teknologi Mataram

DOI:

https://doi.org/10.33480/pilar.v21i1.6485

Keywords:

data augmentation, deep learning, fine-tuning, ResNet50V2, Songket pattern classification

Abstract

This study aims to develop a classification system for traditional Lombok songket fabric patterns using the ResNet50V2 architecture, optimized through fine-tuning and the AdamW optimizer. The data were collected directly from songket artisans in Lombok and categorized into three groups based on the origin of the patterns: Sade, Sukarara, and Pringgasela. The model was trained with data augmentation techniques, including rotation, shifting, and zooming, to increase data diversity. During the training process, fine-tuning was applied to the last layer of ResNet50V2, and optimization was performed using AdamW with a learning rate of 0.0001. The model was evaluated using a confusion matrix, classification report, and analysis of accuracy and loss. The experimental results showed that the model achieved 100% accuracy at the 15th epoch. Furthermore, experiments with different parameters (epochs, batch size, and learning rate) demonstrated that the 15th epoch provided the best results with 100% accuracy, while using higher epochs (30 and 40) did not necessarily yield better outcomes. This model is effective in identifying songket fabric patterns with good classification results for each class. Although the results are excellent, increasing the dataset size and exploring more complex model architectures could further enhance performance. Overall, this study demonstrates the significant potential of deep learning technology in classifying songket patterns with reliable accuracy in real-world applications.

Downloads

Download data is not yet available.

References

Al Sasongko, S. M., Jayanti, E. D., & Ariessaputra, S. (2022). Application of Gray Scale Matrix Technique for Identification of Lombok Songket Patterns Based on Backpropagation Learning. International Journal on Informatics Visualization, 6(4), 835–841. https://doi.org/10.30630/joiv.6.4.1532

Das, S., S, S., M, A., & Jayaram, S. (2021). Deep Learning Convolutional Neural Network for Defect Identification and Classification in Woven Fabric. Indian Journal of Artificial Intelligence and Neural Networking, 1(2), 9–13. https://doi.org/10.54105/ijainn.b1011.041221

Dogan, Y. (2023). A New Global Pooling Method for Deep Neural Networks: Global Average of Top-K Max- Pooling. Traitement Du Signal, 40(2), 577–587. https://doi.org/10.18280/ts.400216

Duklan, N., Kumar, S., Maheshwari, H., Singh, R., Sharma, S. D., & Swami, S. (2024). CNN Architectures for Image Classification : A Comparative Study Using ResNet50V2 , ResNet152V2 ,. SSRG International Journal of Electronics and Communication Engineering, 11(9), 11–21.

Gültekin, E., Çelik, H. İ., & Kaynak, H. K. (2021). a New Approach for Classification of Different Woven Fabric Patterns and Thread Densities With Convolutional Neural Networks. TEXTEH Proceedings, 2021(December), 88–93. https://doi.org/10.35530/tt.2021.10

Hambali, H., & Mahayadi, M. (2021). Classification of Lombok Songket Cloth Image Using Convolution Neural Network Method (Cnn). Pilar Nusa Mandiri, 17(85), 149–156. https://doi.org/10.33480/pilar.v17i2.2705

He, Y., Zhang, H.-D., Huang, X.-Y., & Hock Tay, F. E. (2021). Fabric Defect Detection based on Improved Faster RCNN. International Journal of Artificial Intelligence & Applications, 12(04), 23–32. https://doi.org/10.5121/ijaia.2021.12402

Hindarto, D. (2023). Use ResNet50V2 Deep Learning Model to Classify Five Animal Species. Jurnal JTIK (Jurnal Teknologi Informasi Dan Komunikasi), 7(4), 758–768. https://doi.org/10.35870/jtik.v7i4.1845

Hussain, M. A. I., Khan, B., Wang, Z., & Ding, S. (2020). Woven fabric pattern recognition and classification based on deep convolutional neural networks. Electronics (Switzerland), 9(6), 1–12. https://doi.org/10.3390/electronics9061048

Ilahi, R., B, F. B., Dwiyansaputra, R., & Farinda, R. (2022). Proceedings of the First Mandalika International Multi-Conference on Science and Engineering 2022, MIMSE 2022 (Informatics and Computer Science). In Proceedings of the First Mandalika International Multi-Conference on Science and Engineering 2022, MIMSE 2022 (Informatics and Computer Science) (Vol. 1). Atlantis Press International BV. https://doi.org/10.2991/978-94-6463-084-8

Imran, B., & Efendi, M. M. (2020). THE IMPLEMENTATION OF EXTRACTION FEATURE USING GLCM AND BACK-PROPAGATION ARTIFICIAL NEURAL NETWORK TO CLASIFY LOMBOK SONGKET WOVEN CLOTH. Jurnal Techno Nusa Mandiri, 17(2), 131–136.

Kahraman, Y., & Durmu¸so˘glu, A. (2022). Classification of Defective Fabrics Using Capsule Networks. Appied Sciences, 124–134. https://doi.org/10.4324/9781315232140-14

Kartika, D., & Rahmawati Suwanto, F. (2022). Generating of North Sumatran Songket Fabric Motifs Based on Crystallographic Groups Using MATLAB GUI. ICIESC, January 2022. https://doi.org/10.4108/eai.11-10-2022.2325312

Kim, B. J., Choi, H., Jang, H., Lee, D., & Kim, S. W. (2023). How to Use Dropout Correctly on Residual Networks with Batch Normalization. Proceedings of Machine Learning Research, 216(Uai), 1058–1067.

Lesiangi, F. S., Mauko, A. Y., & Djahi, B. S. (2021). Feature extraction Hue, Saturation, Value (HSV) and Gray Level Cooccurrence Matrix (GLCM) for identification of woven fabric motifs in South Central Timor Regency. Journal of Physics: Conference Series, 2017(1). https://doi.org/10.1088/1742-6596/2017/1/012010

Pei, X., Zhao, Y. hong, Chen, L., Guo, Q., Duan, Z., Pan, Y., & Hou, H. (2023). Robustness of machine learning to color, size change, normalization, and image enhancement on micrograph datasets with large sample differences. Materials and Design, 232, 112086. https://doi.org/10.1016/j.matdes.2023.112086

Tasnim, N. H., Afrin, S., Biswas, B., Anye, A. A., & Khan, R. (2023). Automatic classification of textile visual pollutants using deep learning networks. Alexandria Engineering Journal, 62, 391–402. https://doi.org/10.1016/j.aej.2022.07.039

Welamo1, M. T., & Deng1Sanpeng. (2021). Application of Artificial Intelligence in Textile Industry. Journal of Emerging Technologies and Innovative Research, 8(10), 49–64. www.jetir.orgd49

Yildiz, K. (2020). Identification of wool and mohair fibres with texture feature extraction and deep learning. IET Image Processing, 14(2), 348–353. https://doi.org/10.1049/iet-ipr.2019.0907

Yuhandri, Madenda, S., Wibowo, E. P., & Karmilasari. (2017a). Object feature extraction of songket image using chain code algorithm. International Journal on Advanced Science, Engineering and Information Technology, 7(1), 235–241. https://doi.org/10.18517/ijaseit.7.1.1479

Yuhandri, Madenda, S., Wibowo, E. P., & Karmilasari. (2017b). Pattern recognition and classification using backpropagation neural network algorithm for songket motifs image retrieval. International Journal on Advanced Science, Engineering and Information Technology, 7(6), 2343–2349. https://doi.org/10.18517/ijaseit.7.6.2200

Yullyana, Y., Irmayani, D., & Hasibuan, M. N. S. (2022). Content-Based Image Retrieval for Songket Motifs using Graph Matching. Sinkron, 7(2), 714–719. https://doi.org/10.33395/sinkron.v7i2.11411