Mango Leaf Disease Detection using Threshold with CNN ResNet50 Architecture

Aditya Dwi Baginda, Alfiah Fajriani, Rifa Atus Shalihah

Abstract


Mango leaf diseases pose a significant threat to farmers’ productivity in Indonesia due to the difficulty and inaccuracy of manual diagnosis. A mango leaf disease detection system was developed by optimizing the decision threshold for classification using a ResNet50 Convolutional Neural Network (CNN). The Kaggle dataset consisted of 3,979 mango leaf images across eight classes: healthy, anthracnose, bacterial canker, gall midge, cutting weevil, dieback, sooty mold, and powdery mildew. The raw dataset was processed in Roboflow with an 80:10:10 train-validation-test split, and threefold data augmentation on the training set produced a total of 9,600 images. Decision threshold optimization using the precision-recall curve analysis identified 0.85 as the optimal threshold. At this threshold, precision reached 97.03%, while recall was 94.36%. These results provide a critical reference for agricultural applications in Indonesia, particularly considering local characteristics. The model achieved an F1-score of 95.49% after validation on the augmented dataset specifically tailored for tropical conditions.

Keywords


mango leaf disease detection; decision threshold; CNN ResNet50; data augmentation; agriculture

Full Text:

PDF

References


S. I. Ahmed et al., “MangoLeafBD: A Comprehensive Image Dataset to Classify Diseased and Healthy Mango Leaves,” Data Br., Vol. 47, p. 108941, 2023, DOI: 10.1016/j.dib.2023.108941.

J. R. Hidaya and Jemakmum, “Implementasi Klasifikasi Citra berbasis Tensorflow untuk mendeteksi Penyakit Tanaman pada Aplikasi Agroscan,” J. Fasilkom, Vol. 15, No. 1, pp. 124–130, 2025, DOI: 10.37859/jf.v15i1.8536.

S. Khandelwal, A. Raut, H. Vyawahare, D. Theng, and S. Dhande, “Optimizing Performance in Mango Plant Leaf Disease Classification through Advanced Machine Learning Techniques,” Eng. Technol. Appl. SCI. Res., Vol. 14, No. 6, pp. 18476–18480, 2024, DOI: 10.48084/etasr.8220.

O. Rainio et al., “Comparison of Thresholds for a Convolutional Neural Network Classifying Medical Images,” Int. J. Data SCI. Anal., Vol. 20, No. 3, pp. 2093–2099, 2025, DOI: 10.1007/s41060-024-00584-z.

S. Jayashree and V. Sumalatha, “Plant Leaf Disease Detection using ResNet-50 based on Deep Learning BT - Proceedings of the International Conference on Digital Transformation in Business: Navigating the New Frontiers Beyond Boundaries (DTBNNF 2024),” Atlantis Press, 2024, pp. 150–166. DOI: 10.2991/978-94-6463-433-4_12.

H. Rogers, B. De La Iglesia, T. Zebin, G. Cielniak, and B. Magri, “Advancing Precision Agriculture: Domain-Specific Augmentations and Robustness Testing for Convolutional Neural Networks in Precision Spraying Evaluation,” Neural Comput. Appl., Vol. 36, No. 32, pp. 20211–20229, 2024, DOI: 10.1007/s00521-024-10142-0.

A. Setiawan, A. Nursafitri, E. V. Afnarista, H. Mulyasari, V. H. Zahwa, and Z. Suryadi, “Klasifikasi Penyakit Tanaman Mangga menggunakan Algorithma CNN pada Citra Daun Daun,” Sentimeter. (Seminar Nas. Teknol. Informasi, Mekatronika dan Ilmu Komputer), Vol. 4, pp. 30–41, 2025.

N. C. Rasyid, J. Karman, A. T. Hidayat, H. Oktavia, and L. Wijaya, “Klasifikasi Jenis Penyakit Buah Mangga berbasis Deep Learning menggunakan Arsitektur ResNet dan MobileNet,” No. 31626, pp. 166–176, 2026.

F. I. Tanesab, C. Y. Laatrehe, and M. A. Wuarlela, “Penerapan Metode CNN dalam mengklasifikasi Jenis Penyakit pada Daun Mangga menggunakan Arsitektur InceptionV3,” Vol. 1, pp. 265–275, 2025.

S. T. R. Siagian, I. Taufik, Z. Indra, “Klasifikasi Penyakit dan Hama pada Daun Mangga menggunakan Metode Convolutional Neural Network,” JATI (Jurnal Mhs. Tek. Inform., Vol. 9, No. 3, pp. 4135–4143, 2025, DOI: 10.36040/jati.v9i3.13571.

R. Ardiansyah, M. A. D. Widyadara, and U. Mahdyah, “Deteksi Penyakit Daun Mangga menggunakan Convolutional Neural Network untuk Analisis Komperasi Arsitektur VGG16, Xception,” Inotek, Vol. 9, pp. 2549–7952, 2025.

E. A. Rosyidah, A. F. Hadi, and Y. S. Dewi, The Classification of Tea Leaf Disease using CNN Image Classifier. Atlantis Press International BV, 2023. DOI: 10.2991/978-94-6463-174-6_10.

E. Suherman, B. Rahman, D. Hindarto, and H. Santoso, “Implementation of ResNet-50 on End-to-End Object Detection (DETR) on Objects,” SinkrOn, Vol. 8, No. 2, pp. 1085–1096, 2023, DOI: 10.33395/sinkron.v8i2.12378.

M. A. Akif, I. Butun, A. Williams, and I. Mahgoub, “Hybrid Machine Learning Models for Intrusion Detection in IoT: Leveraging a Real-World IoT Dataset,” pp. 1–9, 2025, [Online]. Available: http://arxiv.org/abs/2502.12382

N. IBRAHIM et al., “Klasifikasi Tingkat Kematangan Pucuk Daun Teh menggunakan Metode Convolutional Neural Network,” ELKOMIKA J. Tek. Energi Elektr. Tek. Telekomun. Tek. Elektron., Vol. 10, No. 1, p. 162, 2022, DOI: 10.26760/elkomika.v10i1.162.

S. A. Adzkia and T. Arifin, “Perbandingan Arsitektur CNN berbasis Transfer Learning untuk Klasifikasi pada BreastMNIST,” J. Nas. Teknol. dan Sist. Inf., Vol. 11, No. 2, pp. 192–200, 2025, DOI: 10.25077/teknosi.v11i2.2025.192-200.

K. F. Ramadhani and M. Tarigan, “Implementasi Metode Convolutional Neural Network untuk Klasifikasi Penyakit Daun Mangga menggunakan Arsitektur Efficientnetv2-S dan ResNet50,” JATI (Jurnal Mhs. Tek. Inform., Vol. 9, No. 3, pp. 4135–4143, 2025, DOI: 10.36040/jati.v9i3.13571.

J. E. Adio et al., “Development of an Optimized Deep Learning Technique for Fabric Defect Classification using Osprey Optimization Algorithm,” FUOYE J. Eng. Technol., Vol. 9, No. 4, pp. 643–648, 2025, DOI: 10.4314/fuoyejet.v9i4.12.




DOI: https://doi.org/10.32520/stmsi.v15i2.6123

Article Metrics

Abstract view : 5 times
PDF - 0 times

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.