Klasifikasi Spesies Nyamuk Berbasis Few-Shot Learning Prototypical Network dengan ResNet-34 untuk Mendukung Sistem Pengendalian Vektor
DOI:
https://doi.org/10.52436/1.jpti.1356Keywords:
Few-Shot Learning, Grad-CAM, Klasifikasi Spesies Nyamuk, Prototypical Network, ResNet-34Abstract
Klasifikasi spesies nyamuk secara cepat dan akurat merupakan aspek penting dalam upaya pengendalian berbagai penyakit seperti demam berdarah, chikungunya, dan filariasis. Metode klasifikasi berbasis pembelajaran mesin konvensional umumnya membutuhkan dataset berukuran besar yang relatif sulit untuk didapatkan. Untuk mengatasi kendala tersebut, penelitian ini mengusulkan pendekatan Few-Shot Learning (FSL) dengan menggunakan arsitektur Prototypical Network yang didukung oleh deep visual embeddings berbasis backbone ResNet-34. Model dilatih secara episodik dengan sedikit data per kelas menggunakan citra dari tiga spesies nyamuk utama yaitu Aedes aegypti, Aedes albopictus, dan Culex quinquefasciatus. Evaluasi kinerja model dilakukan dengan menggunakan metrik akurasi, presisi, recall, dan F1-score. Hasil penelitian ini menunjukkan bahwa model dapat mencapai akurasi rata-rata sebesar 96,33% dengan deviasi antar-fold yang rendah serta stabilitas dan kemampuan generalisasi yang tinggi. Selain akurat, model ini juga efisien secara komputasi dengan waktu pelatihan rata-rata sebesar 0,83 detik per episode. Visualisasi menggunakan Grad-CAM menunjukkan bahwa model secara konsisten dapat memfokuskan perhatian pada area morfologis penting seperti toraks dan abdomen sehingga meningkatkan interpretabilitas dari proses klasifikasi. Secara keseluruhan, penelitian ini memberikan kontribusi penting dalam pengembangan sistem surveilans vektor berbasis kecerdasan buatan di wilayah dengan keterbatasan data dan sumber daya.
Downloads
References
L. Facchinelli, A. Badolo, and P. J. McCall, “Biology and Behaviour of Aedes aegypti in the Human Environment: Opportunities for Vector Control of Arbovirus Transmission,” Mar. 01, 2023, MDPI. doi: 10.3390/v15030636.
N. Yurayart, M. Kaewthamasorn, and S. Tiawsirisup, “Vector competence of Aedes albopictus (Skuse) and Aedes aegypti (Linnaeus) for Plasmodium gallinaceum infection and transmission,” Vet Parasitol, vol. 241, pp. 20–25, Jul. 2017, doi: 10.1016/J.VETPAR.2017.05.002.
M. A. Saleem and I. Lobanova, “Mosquito-borne diseases,” Dengue Virus Disease: From Origin to Outbreak, pp. 57–83, Jan. 2020, doi: 10.1016/B978-0-12-818270-3.00005-9.
M. Gómez, D. Martinez, M. Muñoz, and J. D. Ramírez, “Aedes aegypti and Ae. albopictus microbiome/virome: new strategies for controlling arboviral transmission?,” Parasit Vectors, vol. 15, no. 1, p. 287, 2022, doi: 10.1186/s13071-022-05401-9.
J. Roberts, K. H. Kenyon, A. Kwarteng, and E. K. A. Amewu, “Lymphatic filariasis as an indigenous illness: the local context of a global disease,” SSM - Qualitative Research in Health, vol. 8, p. 100598, Dec. 2025, doi: 10.1016/J.SSMQR.2025.100598.
J. I. Blanford, “Managing vector-borne diseases in a geoAI-enabled society. Malaria as an example,” Acta Trop, vol. 260, p. 107406, Dec. 2024, doi: 10.1016/J.ACTATROPICA.2024.107406.
T. Chaiphongpachara et al., “Geometric morphometric and molecular techniques for discriminating among three cryptic species of the Anopheles barbirostris complex (Diptera: Culicidae) in Thailand,” Heliyon, vol. 8, no. 10, Oct. 2022, doi: 10.1016/j.heliyon.2022.e11261.
K. Nolte, F. G. Sauer, J. Baumbach, P. Kollmannsberger, C. Lins, and R. Lühken, “Robust mosquito species identification from diverse body and wing images using deep learning,” Parasit Vectors, vol. 17, no. 1, p. 372, 2024, doi: 10.1186/s13071-024-06459-3.
S. Lee, H. Kim, and B.-K. Cho, “Deep Learning-Based Image Classification for Major Mosquito Species Inhabiting Korea,” Insects, vol. 14, no. 6, 2023, doi: 10.3390/insects14060526.
A. A. J. Karim, M. Z. Mahmud, and R. Khan, “Advanced vision transformers and open-set learning for robust mosquito classification: A novel approach to entomological studies,” PLoS Comput Biol, vol. 20, no. 12, pp. e1012654-, Dec. 2024, [Online]. Available: https://doi.org/10.1371/journal.pcbi.1012654
E. Pachetti and S. Colantonio, “A systematic review of few-shot learning in medical imaging,” Artif Intell Med, vol. 156, p. 102949, Oct. 2024, doi: 10.1016/J.ARTMED.2024.102949.
J. Snell, K. Swersky, and R. Zemel, “Prototypical networks for few-shot learning,” in Proceedings of the 31st International Conference on Neural Information Processing Systems, in NIPS’17. Red Hook, NY, USA: Curran Associates Inc., 2017, pp. 4080–4090.
B. P. Amiruddin and R. E. Abdul Kadir, “CNN Architectures Performance Evaluation for Image Classification of Mosquito in Indonesia,” in 2020 International Seminar on Intelligent Technology and Its Applications (ISITIA), 2020, pp. 223–227. doi: 10.1109/ISITIA49792.2020.9163732.
S. Lee, H. Kim, and B.-K. Cho, “Deep Learning-Based Image Classification for Major Mosquito Species Inhabiting Korea,” Insects, vol. 14, no. 6, 2023, doi: 10.3390/insects14060526.
Z. Jiang, Z. Feng, and B. Niu, “Prototype-Neighbor Networks with task-specific enhanced meta-learning for few-shot classification,” Neural Networks, vol. 190, p. 107761, Oct. 2025, doi: 10.1016/J.NEUNET.2025.107761.
G. I??k and ?. Paçal, “Few-shot classification of ultrasound breast cancer images using meta-learning algorithms,” Neural Comput Appl, vol. 36, no. 20, pp. 12047–12059, Jul. 2024, doi: 10.1007/s00521-024-09767-y.
J. N. Eckardt et al., “Deep learning identifies Acute Promyelocytic Leukemia in bone marrow smears,” BMC Cancer, vol. 22, no. 1, Dec. 2022, doi: 10.1186/s12885-022-09307-8.
Ong Song Quan, “Mosquito on human skin,” Mendeley Data, Jun. 2022, doi: 10.17632/zw4p9kj6nt.2.
N. Ragu and J. Teo, “Pest classification: Explainable few-shot learning vs. convolutional neural networks vs. transfer learning,” Sci Afr, vol. 27, p. e02512, Mar. 2025, doi: 10.1016/J.SCIAF.2024.E02512.
