A Comparative Evaluation Of Deep Learning Architectures For Butterfly Species Classification In Natural Scene Images
DOI:
https://doi.org/10.29284/mssmkk71Keywords:
Butterfly Species classification, Natural Scene Images, Segmented and Non-Segmented butterfly images, Deep Learning, Transfer Learning, Biodiversity Monitoring.Abstract
Automated ecological species identification is a subject of intense study in computer vision research, which plays an important role in biodiversity monitoring, research, and conservation planning. The research work presented in this article introduces a large custom butterfly dataset consisting of 23537 images belonging to 102 different species, collected through extensive field photography and supplemented with selected images from reliable online sources. Furthermore, the dataset is preprocessed with annotations to generate a Region of Interest (RoI). A total of 54 state-of-the-art deep learning architectures, including classical Convolution Neural Networks (CNN), Modern CNN, Lightweight CNN models, attention-based models, and Vision Transformers, were evaluated on both segmented and non-segmented images of the dataset. A comparative analysis of all the deep learning models in terms of accuracy, precision, recall and F1-Score has been done. The non-uniform class distribution in the dataset motivates the use of weighted evaluation metrics along with overall accuracy, ensuring an impartial evaluation across all categories. Results indicate that modern CNN architectures consistently outperform other models on both segmented and non-segmented images in the dataset. In particular, ConvNeXt-Tiny achieved the best performance of a weighted F1-Score of 90.88% on segmented images and 84.8% on non-segmented images. This proves the effectiveness of object-focused preprocessing and demonstrates the superiority of modern CNN architectures for fine-grained butterfly species classification.
References
1. Josiah Wang and Katja Markert and Mark Everingham, Learning Models for Object Recognition from Natural Language processing Description, “Proceedings of the British Machine Vision Conference” year =”2009”.
2. Seung-Ho Kang, Su-Hee Song, Sang-Hee Lee, Identification of butterfly species with a single neural network system, Journal of Asia-Pacific Entomology 15 (2012) 431–435.
3. Nur Nabila Kamaron Arzar, Nurbaity Sabri, Nur Farahin Mohd Johari, Anis Amilah Shari, MohdRahmat Mohd Noordin, Shafaf Ibrahim, Butterfly Species Identification Using Convolutional Neural Network (CNN), 2019 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS 2019), 29 June 2019, Selangor, Malaysia.
4. Ruoyan Zhao, Cuixia Li1, Shuai Ye and Xinru Fang, Butterfly Recognition Based on Faster R-CNN, IOP Conf. Series: Journal of Physics: Conf. Series 1176 (2019) 032048, doi:10.1088/1742-6596/1176/3/032048.
5. H.Theivaprakasham, Identification of Indian butterflies using Deep ConvolutionalNeural Network, Journal of Asia-Pacific Entomology (2020),doi: https://doi.org/10.1016/j.aspen.2020.11.015.
6. Aijiao tan, guoxiong zhou , and mingfang he, Rapid Fine-Grained Classification of Butterflies Based on FCM-KM and Mask R-CNN Fusion, 20, 2020.Digital Object Identifier 10.1109/ACCESS.2020.3007745.
7. Ayad Saad Almryad, Hakan Kutucu, Automatic identification for field butterflies by convolutional neural networks, Engineering Science and Technology, an International Journal 23 (2020) 189–195.
8. Fathimathul Rajeena P. P., Rasha Orban, Kogilavani Shanmuga Vadivel, Malliga Subramanian, Suresh Muthusamy, Diaa Salam Abd Elminaam, Ayman Nabil, Laith Abulaigh, Mohsen Ahmadi and Mona A. S. Ali. A Novel Method for the Classification of Butterfly Species Using Pre-Trained CNN Models, Electronics 2022, 11, 2016. https://doi.org/10.3390/electronics11132016.
9. Halit Çetiner, A Novel Method for Classification of Butterfly Species Using CNN, 2nd International Conference on Scientific and Academic Research, 2023, Published by All Sciences Proceedings.
10. Rumana Yasmin, Aparna Das, Liton Jude Rozario, Md. Ezharul Islam, Butterfly detection and classification techniques: A review, Intelligent Systems with Applications 18 (2023) 200214.
11. Diniati Ruaika, Shofwatul ’Uyun, Optimisation of Residual Network Using Data Augmentation and Ensemble Deep Learning for Butterfly Image Classification, IJID (International Journal on Informatics for Development), e-ISSN: 2549-7448 Vol. 12, No. 2, December 2023, Pp. 350-361.
12. Abdul Fadlil, Ainin Maftukhah, Sunardi and Tole Sutikno, Butterfly Image Identification Using Multilevel Thresholding,Segmentasi and Convolution Neural Network Classification with Alexnet Architecture, International Journal of Computing and Digital Systems ISSN (2210-142X), dx.doi.org/10.12785/ijcds/1501125.
13. Idir Filali, Mohamed Ramdani, Brahim Achour, Butterfly segmentation by multi scaled quantum cuts in agro-ecological environment, An International Journal, A publication of the European Association for Signal Processing (EURASIP), doi.org/10.1016/j.sigpro.2024.109420.
14. Anwar, H. Butterflies and benchmarks: a systematic evaluation of deep learning backbones to identify the best model for image-based butterflies species recognition. Pattern Anal Applic 29, 13 (2026). https://doi.org/10.1007/s10044-025-01587-7
15. Kunte, K. 2026. ABOUT US. In Kunte, K., S. Sondhi, and P. Roy (Chief Editors). Butterflies of India, v. 4.12. Indian Foundation for Butterflies.
16. ButterflyCircle. (n.d.). Butterfly images. https://www.butterflycircle.com/
17. iNaturalist. (n.d.). Butterflies of Singapore.
18. iNaturalist. https://www.inaturalist.org/projects/butterflies-of-singapore
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
This work is licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
