IMAGE-BASED IDENTIFICATION OF SUMATRA BUTTERFLY SPECIES USING DEEP LEARNING
DOI:
https://doi.org/10.18623/rvd.v23.6532Abstract
Deep learning has gained momentum in the last decade for image-based species identification. We examine the classification of eight Sumatra butterfly species with a dataset of 800 photos from Gita Persada Butterfly Park using several deep learning architectures. The dataset was separated into training, validation, and testing subsets under controlled experimental circumstances. We evaluated seven architectures employing transfer learning with ImageNet pretrained weights, including convolutional neural networks (CNNs) and a Vision Transformer (ViT). The DenseNet201 model obtained the highest classification accuracy (99.38%), followed by ResNet50 and Xception (98.75%), MobileNet (97.50%), InceptionV3 (95.63%), ViT (93.75%), and EfficientNetB0 (85.63%). The performances of CNN-based models were more stable under the conditions of the present dataset, while MobileNet reached a good trade-off between accuracy and computational efficiency. The results must be viewed from the perspective of a tiny, single-site data collection. This was a controlled study and not a deployable field application as a standard for the identification of butterfly species from images. There was no external validation or field-based assessment. The results, therefore, give initial support for the possibility of deep learning for automated identification of butterfly species, but further validation using bigger and more diverse datasets is needed to determine model generalizability.
References
Adityawan, H. T., Farroq, O., Santosa, S., Islam, H. M. M., Sarker, M. K., & Setiadi, D. R. I. M. (2023). Butterflies Recognition using Enhanced Transfer Learning and Data Augmentation. Journal of Computing Theories and Applications, 1(2), 115–128.
Ali, K., Shaikh, Z. A., Khan, A. A., & Laghari, A. A. (2022). Multiclass skin cancer classification using EfficientNets – a first step towards preventing skin cancer. In Neuroscience Informatics, 2(4).
Almryad, A. S., & Kutucu, H. (2020). Automatic identification for field butterflies by convolutional neural networks. Engineering Science and Technology, an International Journal, 23(1), 189–195.
Alzubaidi, L., Zhang, J., Humaidi, A. J., Al-Dujaili, A., Duan, Y., Al-Shamma, O., Santamaría, J., Fadhel, M. A., Al-Amidie, M., & Farhan, L. (2021). Review of deep learning: concepts, CNN architectures, challenges, applications, future directions. Journal of Big Data, 8(1), 53.
Fathimathul, R. P. P., Orban, R., Vadivel, K. S., Subramanian, M., Muthusamy, S., Elminaam, D. S. A., Nabil, A., Abulaigh, L., Ahmadi, M., & Ali, M. A. S. (2022). A Novel Method for the Classification of Butterfly Species Using Pre-Trained CNN Models. Electronics (Switzerland), 11(13).
Hasan, M. R., Rahman, M. M., Shahriar, F., Khan, M. S. I., Mohi Uddin, K. M., & Hasan, M. M. (2024). Smart farming: Leveraging IoT and deep learning for sustainable tomato cultivation and pest management. Crop Design, 3(4).
Ilhamdi, M. L., Al Idrus, A., & Santoso, D. (2018). Diversity of Species and Conservation Priority of Butterfly at Suranadi Natural Park of West Lombok, Indonesia. Biosaintifika, 10(1), 48–55.
Karim, A. A. J., Mahmud, M. Z., & Khan, R. (2024). Advanced vision transformers and open-set learning for robust mosquito classification: A novel approach to entomological studies. PLOS Computational Biology, 20(12).
Koneri, R., Maabuat, P. V., & Nangoy, M. J. (2020). The distribution and diversity of butterflies (Lepidoptera: Rhopalocera) in various urban forests in north Minahasa regency, north Sulawesi province, Indonesia. Applied Ecology and Environmental Research, 18(2), 2295–2314.
Liang, B., Wu, S., Xu, K., & Hao, J. (2020). Butterfly Detection and Classification Based on Integrated YOLO Algorithm, Genetic and Evolutionary Computing (500–512).
Ong, S. Q., & Hamid, S. A. (2022). Next generation insect taxonomic classification by comparing different deep learning algorithms. PLoS ONE, 17(12).
Peggie, D., Prabowo, S. W. B., Shahroni, A. M., Shidiq, F. I. A., Irwansyah, L., Soenarko, S., Rahma, N., & Wafa, I. Y. (2025). Kupunesia App For Citizen Science: New Way of Mainstreaming Interest and Study of Indonesian Butterflies. Treubia, 49(2), 137–148.
Pradnyatama, M., Sari, C. A., Rachmawanto, E. H., & Islam, H. M. M. (2025). A Comparative Analysis of Convolutional Neural Network (CNN): MobileNetV2 and Xception for Butterfly Species Classification. Jurnal Masyarakat Informatika, 16(1), 69–90.
Pucci, R., Kalkman, V. J., & Stowell, D. (2023). Comparison between transformers and convolutional models for fine-grained classification of insects.
Pucci, R., Kalkman, V. J., & Stowell, D. (2025). Performance of Computer Vision Algorithms for Fine-Grained Classification Using Crowdsourced Insect Images. IET Computer Vision, 19(1).
Sagar, V., Sachin, R., Chandrashekara, K., & Ganeshaiah, K. (2020). Identification of Indian Butterflies and Moths with Deep Convolutional Neural Networks x. Current Science, 118, 1456.
Spiesman, B. J., Gratton, C., Hatfield, R. G., Hsu, W. H., Jepsen, S., McCornack, B., Patel, K., & Wang, G. (2021). Assessing the potential for deep learning and computer vision to identify bumble bee species from images. Scientific Reports, 11(1).
Stark, T., Ştefan, V., Wurm, M., Spanier, R., Taubenböck, H., & Knight, T. M. (2023). YOLO object detection models can locate and classify broad groups of flower-visiting arthropods in images. Scientific Reports, 13(1).
Yasmin, R., Das, A., Rozario, L. J., & Islam, M. E. (2023). Butterfly detection and classification techniques: A review. In Intelligent Systems with Applications (18).
Downloads
Published
How to Cite
Issue
Section
License
I (we) submit this article which is original and unpublished, of my (our) own authorship, to the evaluation of the Veredas do Direito Journal, and agree that the related copyrights will become exclusive property of the Journal, being prohibited any partial or total copy in any other part or other printed or online communication vehicle dissociated from the Veredas do Direito Journal, without the necessary and prior authorization that should be requested in writing to Editor in Chief. I (we) also declare that there is no conflict of interest between the articles theme, the author (s) and enterprises, institutions or individuals.
I (we) recognize that the Veredas do Direito Journal is licensed under a CREATIVE COMMONS LICENSE.
Licença Creative Commons Attribution 3.0
