Klasifikasi Hierarkis dengan Kalibrasi Probabilitas untuk Penyakit Daun Singkong Menggunakan Fusi Fitur dan Random Forest
-
Regi Fernando Najoan
*
Universitas Klabat, Minahasa Utara,
Indonesia
- Enrico Samuel Djimesha Universitas Klabat, Minahasa Utara, Indonesia
- Argha Orion Silitonga Universitas Klabat, Minahasa Utara, Indonesia
- Wilsen Grivin Mokodaser Universitas Klabat, Minahasa Utara, Indonesia
- George William Tangka Universitas Klabat, Minahasa Utara, Indonesia (*) Corresponding Author
Abstract
Field-image classification of cassava leaf disease requires trustworthy probabilities and a mechanism for withholding uncertain decisions, rather than label accuracy alone. This study evaluates HSV, GLCM, and MobileNetV3-Small embedding fusion in a two-stage hierarchical Random Forest with probability calibration and selective prediction. A total of 7,508 CCMT images were evaluated using outer stratified 5-fold cross-validation. Model comparisons report mean ± standard deviation across five outer folds. End-to-end MobileNetV3-Small achieved 93.90±0.73% accuracy, 94.64±0.59% balanced accuracy, and 94.23±0.73% Macro F1, exceeding the Hierarchical Hybrid Random Forest at 88.28±1.12%, 89.66±0.76%, and 88.79±1.07%, respectively. The 1.03-percentage-point Macro F1 difference between the hierarchical and Flat Hybrid models was not significant at the 5% level (two-sided Wilcoxon, p=0.125). A separate calibration and selective-prediction experiment divided each outer-train fold into 70% model-fit, 15% calibrator-fit, and 15% threshold-selection partitions, all disjoint from the outer-test fold. Simplex-temperature yielded the lowest ECE (0.54%), whereas sigmoid performed better on log loss and Brier score. Thresholds selected on the held-out threshold-selection partition yielded 75.87% outer-test coverage, 95.40% accepted accuracy, and 95.58% accepted Macro F1. Selective accuracy applies only to accepted samples, not all data or guaranteed future performance; the contribution is evaluation of the coverage–risk trade-off, not improved intrinsic accuracy.
Downloads
References
E. Mwebaze, T. Gebru, A. Frome, S. Nsumba, and J. Tusubira, “iCassava 2019 Fine-Grained Visual Categorization Challenge,” 2019. doi: 10.48550/arXiv.1908.02900.
P. K. Mensah et al., “CCMT: Dataset for Crop Pest and Disease Detection,” Data Brief, vol. 49, p. 109306, Aug. 2023, doi: 10.1016/j.dib.2023.109306.
V. K. Patel, K. Abhishek, and S. Selvarajan, “Optimized Recurrent Neural Network-Based Early Diagnosis of Crop Pest and Diseases in Agriculture,” Discover Computing, vol. 27, no. 1, p. 43, Nov. 2024, doi: 10.1007/s10791-024-09481-2.
B. Jia, Z. Zheng, J. Zeng, L. Shi, H. Ge, and C. Song, “Interpretable Crop Pest and Disease Identification Based on Comparative Concept Tree,” PLoS One, vol. 21, no. 3, p. e0343715, Mar. 2026, doi: 10.1371/journal.pone.0343715.
A. M. Anggraeni, T. I. Hermanto, and I. M. Nugroho, “Klasifikasi Penyakit Daun Singkong Menggunakan Convolutional Neural Network (CNN) dengan Arsitektur VGG16 Berbasis Android,” Jurnal Teknologi Terpadu, vol. 11, no. 1, pp. 53–59, Jul. 2025, [Online]. Available: https://journal.nurulfikri.ac.id/index.php/JTT/article/view/1862
G. A. Defanka, A. Junaidi, and H. Maulana, “Penerapan LBP dan FCH untuk Klasifikasi Penyakit Daun Jagung Menggunakan Algoritma Random Forest,” Jurnal Informatika Teknologi dan Sains (JINTEKS), vol. 8, no. 1, pp. 85–90, Feb. 2026, [Online]. Available: https://jurnal.uts.ac.id/JINTEKS/article/view/6948
A. A. Huda, B. Setiaji, and F. R. Hidayat, “Implementasi Gray Level Co-Occurrence Matrix (GLCM) untuk Klasifikasi Penyakit Daun Padi,” Pseudocode, vol. 9, no. 1, pp. 33–38, Feb. 2022, doi: 10.33369/pseudocode.9.1.33-38.
S. K. Wildah, A. Latif, A. Mustopa, Suharyanto, M. S. Maulana, and A. Sasongko, “Klasifikasi Penyakit Daun Kopi Menggunakan Kombinasi Haralick, Color Histogram dan Random Forest,” JUSTIN (Jurnal Sistem dan Teknologi Informasi), vol. 11, no. 1, pp. 35–39, Jan. 2023, doi: 10.26418/justin.v11i1.60985.
Y. Zhong, B. Huang, and C. Tang, “Classification of Cassava Leaf Disease Based on a Non-Balanced Dataset Using Transformer-Embedded ResNet,” Agriculture, vol. 12, no. 9, p. 1360, Sep. 2022, doi: 10.3390/agriculture12091360.
D. Tran-Anh, Q. Nguyen Huu, T. Nguyen Thi Phuong, and Q. Dao Thi Thuy, “Leaf Disease Classification with Multiple-Model Deep Learning,” Journal of Intelligent & Fuzzy Systems, vol. 46, no. 1, pp. 2811–2823, Jan. 2024, doi: 10.3233/JIFS-235940.
A. Howard et al., “Searching for MobileNetV3,” in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, Korea, 2019, pp. 1314–1324. doi: 10.1109/ICCV.2019.00140.
A. Dosset et al., “Cassava Disease Detection Using a Lightweight Modified Soft Attention Network,” Pest Manag. Sci., vol. 81, no. 2, pp. 607–617, Feb. 2025, doi: 10.1002/ps.8456.
H.-T. Thai, K.-H. Le, and N. L.-T. Nguyen, “FormerLeaf: An Efficient Vision Transformer for Cassava Leaf Disease Detection,” Comput. Electron. Agric., vol. 204, p. 107518, Jan. 2023, doi: 10.1016/j.compag.2022.107518.
C. Guo, G. Pleiss, Y. Sun, and K. Q. Weinberger, “On Calibration of Modern Neural Networks,” in Proceedings of the 34th International Conference on Machine Learning (ICML), PMLR, Sydney, Australia, 2017, pp. 1321–1330. [Online]. Available: https://proceedings.mlr.press/v70/guo17a.html
A. Niculescu-Mizil and R. Caruana, “Predicting Good Probabilities with Supervised Learning,” in Proceedings of the 22nd International Conference on Machine Learning (ICML), Bonn, Germany, 2005, pp. 625–632. doi: 10.1145/1102351.1102430.
Y. Geifman and R. El-Yaniv, “Selective Classification for Deep Neural Networks,” in Advances in Neural Information Processing Systems 30 (NeurIPS), Long Beach, CA, USA, 2017, pp. 4878–4887. [Online]. Available: https://papers.nips.cc/paper/2017/hash/4a8423d5e91fda00bb7e46540e2b0cf1-Abstract.html
S. Goren, I. Galil, and R. El-Yaniv, “Hierarchical Selective Classification,” in Advances in Neural Information Processing Systems 37 (NeurIPS), Vancouver, Canada, 2024, pp. 111047–111073. doi: 10.52202/079017-3526.
R. C. Gonzalez and R. E. Woods, Digital Image Processing, 4th ed. New York, NY, USA: Pearson, 2018.
R. Dijaya, Buku Ajar Pengolahan Citra Digital. Sidoarjo: UMSIDA Press, 2023. doi: 10.21070/2023/978-623-464-075-5.
Scikit-learn Developers, “sklearn.ensemble.RandomForestClassifier,” 2025. [Online]. Available: https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestClassifier.html
S. M. Lundberg and S.-I. Lee, “A Unified Approach to Interpreting Model Predictions,” in Advances in Neural Information Processing Systems 30 (NeurIPS), Long Beach, CA, USA, 2017, pp. 4765–4774. [Online]. Available: https://papers.nips.cc/paper/2017/hash/8a20a8621978632d76c43dfd28b67767-Abstract.html
Bila bermanfaat silahkan share artikel ini
Berikan Komentar Anda terhadap artikel Klasifikasi Hierarkis dengan Kalibrasi Probabilitas untuk Penyakit Daun Singkong Menggunakan Fusi Fitur dan Random Forest
Pages: 68-79
Copyright (c) 2026 Regi Fernando Najoan, Enrico Samuel Djimesha, Argha Orion Silitonga, Wilsen Grivin Mokodaser, George William Tangka
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under Creative Commons Attribution 4.0 International License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (Refer to The Effect of Open Access).
