Comparative Analysis of Transfer Learning-Based Deep Learning Models for Jatropha Leaf Disease Classification

Sarifah Agustiani; Sulistiyah; Agus Junaidi; Cucu Ika Agustyaningrum; Yoseph Tajul Arifin

doi:10.32664/smatika.v16i02.2325

Penulis

Sarifah Agustiani Universitas Bina Sarana Informatika
Sulistiyah Universitas Bina Sarana Informatika
Agus Junaidi Universitas Bina Sarana Informatika
Cucu Ika Agustyaningrum Universitas Bina Sarana Informatika
Yoseph Tajul Arifin Universitas Bina Sarana Informatika

DOI:

https://doi.org/10.32664/smatika.v16i02.2325

Kata Kunci:

Deep Learning, Image Classification, Jatropha, Plant Disease Classification, Transfer Learning

Abstrak

Plant disease identification is essential for enhancing agricultural productivity and promoting sustainable crop management practices. Jatropha curcas has considerable potential as a biofuel-producing plant; however, its growth and productivity can be significantly affected by various leaf diseases. Conventional disease diagnosis often requires substantial time and relies heavily on expert knowledge, creating a need for automated solutions based on deep learning techniques. Although deep learning has been widely applied in plant disease recognition, comparative studies focusing on transfer learning models for Jatropha leaf disease classification remain limited, particularly for datasets characterized by distinctive visual features and relatively small sample sizes. This research conducts a comparative assessment of several deep learning architectures to determine the most effective model for classifying Jatropha leaf diseases. The evaluated architectures include MobileNetV2, EfficientNetB0, ResNet50, DenseNet121, and VGG16. All models utilized ImageNet pre-trained weights and were adapted through fine-tuning of the final classification layers to accommodate a dataset containing healthy and diseased Jatropha leaf images. Experimental findings reveal that ResNet50 achieved the highest classification accuracy of 93.81%, followed by VGG16 at 93.58% and EfficientNetB0 at 90.49%. In comparison, DenseNet121 and MobileNetV2 attained accuracies of 85.40% and 74.56%, respectively. Model effectiveness was assessed using accuracy, training duration, confusion matrix analysis, and ROC curve evaluation to examine classification capability across categories. The results demonstrate that ResNet50 offers the most balanced combination of predictive accuracy and performance stability. Overall, the study confirms that transfer learning-based deep learning models are highly effective for Jatropha leaf disease classification, with ResNet50 emerging as the most suitable architecture among those investigated. These findings may serve as a valuable reference for the development of reliable and efficient plant disease detection systems in agricultural environments.

Referensi

[1] J. Kotwal, D. R. Kashyap, and D. S. Pathan, “Agricultural Plant Diseases Identification: From Traditional Approach to Deep Learning,” Mater. Today Proc., vol. 80, pp. 344–356, 2023, doi: 10.1016/j.matpr.2023.02.370.

[2] T. Raharjo, H. P. Putro, and H. E. Sari, “Early Detection of Tomato Plant Diseases Using the Real-Time Detection Transformer (RT-DETR) Model,” J. Ilm. Sist. Inf., vol. 5, no. 1, pp. 584–592, 2026, doi: 10.51903/286bt086.

[3] Y. Gai and H. Wang, “Plant Disease: A Growing Threat to Global Food Security,” Agronomy, vol. 14, no. 8, 2024, doi: 10.3390/agronomy14081615.

[4] S. Agustiani, R. Aryanti, S. K. Wildah, Y. T. Arifin, S. Marlina, and T. Misriati, “Optimisasi Model Deep Learning untuk Deteksi Penyakit Daun Tebu dengan Fine-Tuning MobileNetV2,” J. Informatics Manag. Inf. Technol., vol. 4, no. 4, pp. 150–157, 2024.

[5] Maftuchah and A. Zainudin, Mengenal Tanaman Jarak Pagar (Jatropha Curcas Linn.). Deepublish, 2018.

[6] I. Maulana and I. Darliana, “Efek Mikoriza Arbuskula (MA) dan Zat Perangsang Tumbuh (ZPT) Akar terhadap Pertumbuhan Vegetatif Bibit Tanaman Jarak Pagar (Jatropha curcas L.) Asal Stek Batang,” Compos. J. Ilmu Pertan., vol. 7, no. 2, pp. 130–144, 2025, doi: 10.37577/composite.v7i2.888.

[7] D. Wardani, “Isolasi, Identifikasi, dan Aktivitas Antibakteri Fungi Endofit Daun Jarak Pagar (Jatropha Curcas L) terhadap Klebsiella pneumoniae dan Streptococcus pneumoniae,” J. Mhs. Ilmu Farm. dan Kesehat., vol. 2, no. 1, 2024, doi: 10.59841/jumkes.v2i1.662.

[8] M. Silalahi, “Jatropha curcas L.: Aspek Botani, Potensi Biofuel, Toksisitas, dan Bioaktivitas Marina,” J. Pro-Life, vol. 12, no. 1, pp. 39–50, 2025, doi: 10.33541/pro-life.v12i1.6385.

[9] M. Gusti, T. Ananda, W. Pardosi, A. Maysari, and B. Sembiring, “Literatur Review: Metode Thresholding Untuk Mengidentifikasi Penyakit Pada Daun Bawang Merah,” FIMERKOM J. Inf. Syst. Technol., vol. 1, no. 1, pp. 6–10, 2024.

[10] A. Abbas, S. Jain, M. Gour, and S. Vankudothu, “Tomato Plant Disease Detection Using Transfer Learning with C-GAN Synthetic Images,” Comput. Electron. Agric., vol. 187, p. 106279, 2021, doi: 10.1016/j.compag.2021.106279.

[11] A. T. Adiana, Jumadi, and E. Nurlatifah, “Klasifikasi Penyakit Pada Daun Padi Menggunakan Model Hybrid CNN-SVM,” SMATIKA STIKI Inform. J., vol. 15, no. 2, pp. 258–267, 2025, doi: 10.32664/smatika.v15i02.1548.

[12] K. Anwar, “Sistem Deteksi Wajah Berbasis Deep Learning Menggunakan Convolutional Neural Network (CNN),” J. Comput. Sci. Inf. Technol., vol. 1, no. 2, pp. 46–52, 2025, doi: 10.70716/jocsit.v1i2.258.

[13] S. Agustiani, Y. T. Arifin, A. Junaidi, S. K. Wildah, and A. Mustopa, “Klasifikasi Penyakit Daun Padi Menggunakan Random Forest dan Color Histogram,” vol. 10, no. 1, 2022, doi: 10.23960/komputasi.v10i1.2961.

[14] E. F. Pangestu and B. Irawan, “Klasifikasi Penyakit Tanaman Padi Menggunakan Metode Convolutional Neural Network dengan Arsitektur MobileNetV2,” JITET, vol. 14, no. 1, pp. 1003–1013, 2026, doi: 10.23960/jitet.v14i1.8900.

[15] A. Anis, P. Wang, C. Li, and F. Sohel, “A Survey of Deep Learning Techniques for Image-Based Disease Detection in Dicot Plants,” Inf. Process. Agric., 2026, doi: 10.1016/j.inpa.2026.03.014.

[16] A. I. Khan, S. M. K. Quadri, S. Banday, and J. L. Shah, “Deep Diagnosis: A Real-Time Apple Leaf Disease Detection System Based on Deep Learning,” Comput. Electron. Agric., vol. 198, p. 107093, 2022, doi: 10.1016/j.compag.2022.107093.

[17] S. Siswanto, M. U. Dewi, S. Kholifah, G. Widhiati, and W. Aryani, “Penggunaan Model Deep Learning Untuk Meningkatkan Efisiensi Dalam Aplikasi Machine Learning,” J. Penelit. Sist. Inf., vol. 1, no. 4, pp. 215–238, 2023, doi: 10.54066/jpsi.v1i4.1619.

[18] R. Gunawan and others, “Pendekatan Transfer Learning untuk Klasifikasi Penyakit Mata Menggunakan Citra dengan CNN InceptionV3,” J. Comput. Sci. Inf. Technol., vol. 6, no. 1, pp. 60–67, 2025, doi: 10.37859/coscitech.v6i1.8509.

[19] S. Nabila and others, “Analisi Pengaruh Urutan Preprocessing Terhadap Kinerja MobileNetV2 dan VGG16 untuk Klasifikasi Penyakit Daun,” J. Comput. Sci. Artif. Intell., vol. 6, no. 2, pp. 1–6, 2025, doi: 10.32485/jcsai.JOURNAL.

[20] T. Yunan, A. Rajab, and N. Nafiiyah, “Evaluasi Kinerja Model CNN Berbasis Transfer Learning dalam Klasifikasi Penyakit Daun Padi,” J. Sist. Komput. dan Inform., vol. 7, no. 3, pp. 989–995, 2026, doi: 10.30865/json.v7i3.9539.

[21] I. K. P. G. Hartawan and I. M. G. Sunarya, “Tinjauan Sistematis Literatur Tentang Sistem Deteksi Penyakit Berbasis Deep Learning,” Digit. Transform. Technol., vol. 6, no. 1, pp. 112–123, 2026.

[22] F. Putra, “Penerapan Teknologi Machine Learning dalam Deteksi Dini Penyakit Pada Tanaman Pangan,” J. Kolaborasi Sains dan Ilmu Terap., vol. 3, no. 1, pp. 1–5, 2024.

[23] N. P. Maylianti, I. G. Ngurah, L. Wijayakusuma, I. P. Chandra, and A. Wiguna, “Comparison of EfficientNet-B0 and ResNet-50 for Detecting Diseases in Cocoa Fruit,” J. Appl. Informatics Comput., vol. 9, no. 1, pp. 115–120, 2025, doi: 10.30871/jaic.v9i1.8868.

[24] N. Pratama, A. S. Sunge, and E. Budiarto, “Penerapan Model MobileNetV2 Untuk Prediksi Tingkat Roasting Biji Kopi Berbasis Gambar Pada Bot Telegram,” RIGGS J. Artif. Intell. Digit. Bus., vol. 4, no. 2, pp. 4571–4576, 2025, doi: 10.31004/riggs.v4i2.1272.

[25] T. D. A. R. Putra and Y. F. Riti, “Implementasi Arsitektur MobileNetV2 untuk Klasifikasi Rumah Adat Berbasis Android,” JUSTIN (Jurnal Sist. dan Teknol. Informasi), vol. 14, no. 2, pp. 217–222, 2026.

[26] S. S. Chouhan, U. P. Singh, U. Sharma, and S. Jain, “Leaf Disease Segmentation and Classification of Jatropha Curcas L. and Pongamia Pinnata L. Biofuel Plants Using Computer Vision Based Approaches,” Measurement, vol. 171, p. 108796, 2021, doi: 10.1016/j.measurement.2020.108796.

[27] S. S. K. B, A. M, L. G. D, N. R, P. P, and S. R, “Jatropha Leaf Disease Classification Using Improved Inception ResNetV2,” in 2023 2nd International Conference on Vision Towards Emerging Trends in Communication and Networking Technologies (ViTECoN), 2023, pp. 1–5. doi: 10.1109/ViTECoN58111.2023.10157396.

[28] E. Saxena, S. S. Chouhan, A. Shukla, R. K. Patel, S. Solanki, and U. P. Singh, “Machine Vision Methodology for Jatropha Plant Leaf Disease Detection,” in Trends in Computer Engineering, Data Science and Artificial Intelligence, 2026, pp. 159–168. doi: 10.1007/978-3-032-10199-0_10.

[29] T. H. Saragih, D. M. N. Fajri, A. Hamdianah, W. F. Mahmudy, and Y. P. Anggodo, “Jatropha Curcas Disease Identification Using Fuzzy Neural Network,” in 2017 International Conference on Sustainable Information Engineering and Technology (SIET), 2017, pp. 305–309. doi: 10.1109/SIET.2017.8304153.

[30] S. S. Chouhan, U. P. Singh, A. Kaul, and S. Jain, “A Data Repository of Leaf Images: Practice Towards Plant Conservation with Plant Pathology,” in 2019 4th International Conference on Information Systems and Computer Networks (ISCON), 2019, pp. 700–707. doi: 10.1109/ISCON47742.2019.9036158.

[31] S. A. Maulana, S. H. Batubara, Y. Permata, P. Pasaribu, H. Syahputra, and F. Ramadhani, “Deteksi Burung Menggunakan Convolutional Neural Network (CNN) dengan Model Arsitektur MobileNetV2,” JATI (Jurnal Mhs. Tek. Inform., vol. 8, no. 4, pp. 6108–6114, 2024, doi: 10.36040/jati.v8i4.10126.

[32] T. Berliani, E. Rahardja, and L. Septiana, “Perbandingan Kemampuan Klasifikasi Citra X-ray Paru-paru Menggunakan Transfer Learning ResNet-50 dan VGG-16,” J. Media Heal., vol. 5, no. 2, pp. 123–135, 2023, doi: 10.28932/jmh.v5i2.6116.

[33] D. A. Agustina, “Klasifikasi Citra Jenis Kulit Wajah dengan Algoritma Convolutional Neural Network (CNN) ResNet-50,” J. Ris. Sist. Inf., vol. 1, no. 3, pp. 2–8, 2024, doi: 10.22373/cj.v8i2.25441.

[34] I. Megahaztuti, B. Yuwono, B. M. Akbar, H. C. Rustamaji, and S. P. Nugroho, “Performance Comparison of VGG-19 and DenseNet-121 Architectures for Rice Plant Disease,” Telematika, vol. 22, no. 2, pp. 74–88, 2025, doi: 10.31315/telematika.v22i2.14437.

[35] D. Setiawan, S. Widodo, T. Ridwan, and R. Ambari, “Perancangan Deteksi Emosi Manusia Berdasarkan Ekspresi Wajah Menggunakan Algoritma VGG16,” Syntax J. Inform., vol. 11, no. 1, pp. 1–11, 2022, doi: 10.35706/syji.v11i01.6594.

[36] M. Azhari, Z. Situmorang, and R. Rosnelly, “Perbandingan Akurasi, Recall, dan Presisi Klasifikasi pada Algoritma C4.5, Random Forest, SVM dan Naive Bayes,” J. Media Inform. Budidarma, vol. 5, pp. 640–651, 2021, doi: 10.30865/mib.v5i2.2937.

[37] D. P. Wijaya, L. D. Murti, and M. R. Rachman, “Recall dan Precision pada Online Public Access Catalog (OPAC) Dinas Arsip dan Perpustakaan Kota Bandung,” Visi Pustaka, vol. 24, no. 1, 2022, doi: 10.37014/visipustaka.v24i1.2915.

[38] A. Ridhovan and A. Suharso, “Penerapan Metode Residual Network (ResNet) dalam Klasifikasi Penyakit pada Daun Gandum,” JIPI (Jurnal Ilm. Penelit. dan Pembelajaran Inform., vol. 7, no. 1, pp. 58–65, 2022, doi: 10.29100/jipi.v7i1.2410.

[39] D. Kurniawan, M. Wahyudi, and L. Pujiastuti, “Deteksi dan Prediksi Cerdas Penyakit Paru-Paru dengan Algoritma Random Forest,” Indones. J. Comput. Sci., vol. 3, no. 1, 2024, doi: 10.31294/ijcs.v3i1.6071.

[40] M. C. Rani, F. D. Azkia, R. A. Dewi, M. Wahyudi, Sumanto, and A. S. Budiman, “Perbandingan Algoritma Random Forest, Naive Bayes, dan Neural Network dalam Klasifikasi Penyakit Jantung,” J. Sains Inform. Terap., vol. 4, no. 2, pp. 77–84, 2025, doi: 10.62357/jsit.v4i2.609.