Fine-Tuned YOLOv12 for Robust Cassava Leaf Disease Detection Under Occlusion and Complex Background Conditions
DOI:
https://doi.org/10.68337/cpsm.v1.i1.2026-021Keywords:
YOLOv12, cassava leaf disease, occlusion, complex background, accuracy-efficiency trade-offAbstract
Cassava is a major food commodity in Indonesia, and its production is threatened by various leaf diseases that can reduce yields by up to 95%. Manual identification by farmers is often subjective and time-consuming, while conventional deep learning models frequently suffer a significant drop in accuracy under field conditions with complex backgrounds and occlusion. This study proposes a fine-tuned YOLOv12-based approach for cassava leaf disease detection under challenging field conditions. The dataset was built from two public Kaggle repositories, the main one containing 21,397 images across five classes. A whole-leaf bounding-box annotation strategy was adopted to provide stronger morphological context for detecting disease symptoms under occluded and visually complex conditions. Hyperparameters were optimized by Bayesian optimization with the tree-structured Parzen estimator (TPE) in the Optuna framework. The fine-tuned YOLOv12m model achieved a mean average precision (mAP) of 0.808 at an intersection over union (IoU) threshold of 0.5 (mAP50) and of 0.600 over IoU thresholds from 0.5 to 0.95 (mAP50-95), outperforming the YOLOv12s baseline. Although the computational load increased from 21.5 to 67.8 giga floating-point operations (GFLOPs), the model was considerably better at distinguishing disease features from field background noise. These findings indicate that the proposed approach improves detection performance and resilience for cassava leaf disease detection in complex field environments.
References
[1] A. W. Muhaimin, Hery Toiba, Dwi Retnoningsih, and L. M. Yapanto, "The impact of technology adoption on income and food security of smallholder cassava farmers: Empirical evidence from Indonesia," Int. J. Adv. Sci. Technol., vol. 29, no. 9s, pp. 699-707, 2020. [Online]. Available: http://sersc.org/journals/index.php/IJAST/issue/view/282
[2] Nasir Saleh, Didik Harnowo, and I Made Jana Mejaya, Penyakit-Penyakit Penting pada Ubi Kayu: Deskripsi, Bioekologi, dan Pengendaliannya. Malang, Indonesia: Balai Penelitian Tanaman Aneka Kacang dan Umbi, 2016, ISBN: 978-602-95497-9-9. [Online]. Available: https://repository.pertanian.go.id/handle/123456789/20356
[3] Xuewei Wang, Jun Liu, and Guoxu Liu, "Diseases detection of occlusion and overlapping tomato leaves based on deep learning," Front. Plant Sci., vol. 12, Art. no. 792244, 2021, doi: 10.3389/fpls.2021.792244.
[4] Emmanuel Moupojou, Appolinaire Tagne, Florent Retraint, Anicet Tadonkemwa, Dongmo Wilfried, Hyppolite Tapamo, and Marcellin Nkenlifack, "FieldPlant: A dataset of field plant images for plant disease detection and classification with deep learning," IEEE Access, vol. 11, pp. 35398-35410, 2023, doi: 10.1109/ACCESS.2023.3263042.
[5] Sk Mahmudul Hassan and Arnab Kumar Maji, "Plant disease identification using a novel convolutional neural network," IEEE Access, vol. 10, pp. 5390-5401, 2022, doi: 10.1109/ACCESS.2022.3141371.
[6] Ranjan Sapkota, Zhichao Meng, Martin Churuvija, Xiaoqiang Du, Zenghong Ma, and Manoj Karkee, "Comprehensive performance evaluation of YOLOv12, YOLO11, YOLOv10, YOLOv9 and YOLOv8 on detecting and counting fruitlet in complex orchard environments," Agric. Commun., vol. 4, no. 1, Art. no. 100125, 2026, doi: 10.1016/j.agrcom.2026.100125.
[7] Yanpeng Ji, Tianxiang Ma, Hongliang Shen, Haiyan Feng, Zizi Zhang, Dan Li, and Yuling He, "Transmission line defect detection algorithm based on improved YOLOv12," Electronics, vol. 14, no. 12, Art. no. 2432, 2025, doi: 10.3390/electronics14122432.
[8] A. Chandrashekhar, B. Satyanarayana, Rajani Reddy Gorrepati, Ponduri Vasanthi, and Kothala Lakshmi Prasanna, "An efficient YOLOv12-based framework for detecting extremely small-scale objects," Sci. Rep., vol. 16, Art. no. 2062, 2026, doi: 10.1038/s41598-025-31803-7.
[9] Muhammad Shafay, Taimur Hassan, Muhammad Owais, Irfan Hussain, Sajid Gul Khawaja, Lakmal Seneviratne, and Naoufel Werghi, "Recent advances in plant disease detection: Challenges and opportunities," Plant Methods, vol. 21, Art. no. 140, 2025, doi: 10.1186/s13007-025-01450-0.
[10] Ultralytics, "YOLO hyperparameter tuning," Ultralytics YOLO Docs, 2025. [Online]. Available: https://docs.ultralytics.com/guides/hyperparameter-tuning/
[11] Ultralytics, "Model training with Ultralytics YOLO," Ultralytics YOLO Docs, 2025. [Online]. Available: https://docs.ultralytics.com/modes/train/
[12] Qizhao Chen, "Traffic object detection using YOLOv12," Open Access Libr. J., vol. 12, no. 8, pp. 1-15, 2025, doi: 10.4236/oalib.1113991.
[13] Haitao Wu, Xiaotian Mo, Sijian Wen, Kanglei Wu, Yu Ye, Yongmei Wang, and Youhua Zhang, "DNE-YOLO: A method for apple fruit detection in diverse natural environments," J. King Saud Univ. Comput. Inf. Sci., vol. 36, no. 9, Art. no. 102220, 2024, doi: 10.1016/j.jksuci.2024.102220.
[14] Ultralytics, "What is intersection over union (IoU)?," Ultralytics Glossary, 2025. [Online]. Available: https://www.ultralytics.com/glossary/intersection-over-union-iou
[15] Jierun Chen, Shiu-hong Kao, Hao He, Weipeng Zhuo, Song Wen, Chul-Ho Lee, and S.-H. Gary Chan, "Run, don't walk: Chasing higher FLOPS for faster neural networks," in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), 2023, pp. 12021-12031, doi: 10.1109/CVPR52729.2023.01157.
[16] Laixiang Xu, Wenhao Wang, Zhaopeng Cai, Wenwen Huang, Fengyan Zhai, and Junmin Zhao, "CFNet-YOLOv12: A novel deep learning framework for apple leaf disease detection," Hortic. Plant J., in press, 2026, doi: 10.1016/j.hpj.2025.09.017.
Downloads
Published
Versions
- 2026-10-01 (2)
- 2026-09-30 (1)
Data Availability Statement
This study used the publicly available Cassava Leaf Disease Classification and iCassava 2019 Fine-Grained Visual Categorization Challenge datasets (Kaggle).Conference Proceedings Volume
Section
Categories
License
Copyright (c) 2026 The Authors
This work is licensed under a Creative Commons Attribution 4.0 International License.
Articles published in Conference Proceedings in Science and Management are made available under the Creative Commons Attribution 4.0 International License (CC BY 4.0). Authors retain copyright in their work and grant the journal first publication rights. Readers are free to share and adapt the published material for any purpose, including commercial use, provided appropriate credit is given to the author, a link to the license is provided, and any changes are indicated.
