YOLOv9s with Region-Dispersion Channel Spatial Attention for Robust Chili Leaf Disease Detection
DOI:
https://doi.org/10.62411/jcta.16046Keywords:
Attention mechanism, Computer vision, Deep learning, Plant disease detection, Precision agriculture, Region-Dispersion Channel Spatial Attention, Sustainable agriculture, YOLOv9sAbstract
Abstract: Detecting chili leaf diseases remains challenging due to the non-uniform manifestation of symptoms, local discoloration, small lesion regions, and visual similarity between disease patterns and natural leaf background variations. Although YOLO-based detectors provide favorable computational efficiency, lightweight variants often struggle to distinguish subtle lesion characteristics, while conventional attention mechanisms such as CBAM primarily rely on global feature aggregation and may overlook regional activation variability. To address these limitations, this study proposes a YOLOv9s-based detection framework integrated with a Region-Dispersion Channel Spatial Attention (RDCSA) module. The proposed module incorporates regional dispersion statistics, namely mean, standard deviation, and range, as channel descriptors to capture inter-region feature variability before applying spatial attention refinement. Experiments were conducted on the COLD dataset containing 532 original images from five chili leaf condition categories using a split-before-augmentation protocol to ensure objective evaluation. RDCSA was integrated at the P5 feature level and evaluated through attention placement analysis, component-wise ablation, sensitivity analysis, stability assessment, and comparison with modern attention mechanisms. The proposed YOLOv9s + RDCSA model achieved an mAP@50 of 0.894, mAP@50–95 of 0.773, precision of 0.858, recall of 0.861, and an F1-score of 0.859 with only a marginal increase in model parameters. The results suggest that regional dispersion-based attention improves feature discrimination while preserving computational efficiency, particularly for disease symptoms characterized by heterogeneous spatial patterns. Nevertheless, performance remains influenced by visually ambiguous symptom categories, indicating that further validation across multiple datasets and field conditions is required. Overall, the proposed RDCSA module enhances detection capability without substantially increasing computational overhead, making it a promising attention mechanism for lightweight plant disease detection systems.References
M. P. Aishwarya and A. P. Reddy, “Dataset of chilli and onion plant leaf images for classification and detection,” Data Br., vol. 54, p. 110524, Jun. 2024, doi: 10.1016/j.dib.2024.110524.
A. Jafar, N. Bibi, R. A. Naqvi, A. Sadeghi-Niaraki, and D. Jeong, “Revolutionizing agriculture with artificial intelligence: plant disease detection methods, applications, and their limitations,” Front. Plant Sci., vol. 15, Mar. 2024, doi: 10.3389/fpls.2024.1356260.
T. Nyawose, R. C. Maswanganyi, and P. Khumalo, “A Review on the Detection of Plant Disease Using Machine Learning and Deep Learning Approaches,” J. Imaging, vol. 11, no. 10, p. 326, Sep. 2025, doi: 10.3390/jimaging11100326.
A. Upadhyay et al., “Deep learning and computer vision in plant disease detection: a comprehensive review of techniques, models, and trends in precision agriculture,” Artif. Intell. Rev., vol. 58, no. 3, p. 92, Jan. 2025, doi: 10.1007/s10462-024-11100-x.
K. Kanna S, K. Ramalingam, P. P, J. R, and P. P.C., “YOLO deep learning algorithm for object detection in agriculture: a review,” J. Agric. Eng., vol. 55, no. 4, Dec. 2024, doi: 10.4081/jae.2024.1641.
A. Sharma, V. Kumar, and L. Longchamps, “Comparative performance of YOLOv8, YOLOv9, YOLOv10, YOLOv11 and Faster R-CNN models for detection of multiple weed species,” Smart Agric. Technol., vol. 9, p. 100648, Dec. 2024, doi: 10.1016/j.atech.2024.100648.
T. Y. Mahesh and M. P. Mathew, “Detection of Bacterial Spot Disease in Bell Pepper Plant Using YOLOv3,” IETE J. Res., vol. 70, no. 3, pp. 2583–2590, Mar. 2024, doi: 10.1080/03772063.2023.2176367.
Y. Alhwaiti, M. Khan, M. Asim, M. H. Siddiqi, M. Ishaq, and M. Alruwaili, “Leveraging YOLO deep learning models to enhance plant disease identification,” Sci. Rep., vol. 15, no. 1, p. 7969, Mar. 2025, doi: 10.1038/s41598-025-92143-0.
M. Jelali, “Deep learning networks-based tomato disease and pest detection: a first review of research studies using real field datasets,” Front. Plant Sci., vol. 15, Oct. 2024, doi: 10.3389/fpls.2024.1493322.
C. Gupta et al., “Deep vision in agriculture: assessing the function of YOLO in the classification of plant leaf diseases (PLDs),” BioData Min., vol. 18, no. 1, p. 91, Nov. 2025, doi: 10.1186/s13040-025-00497-y.
J. S. Aguilar-Ruiz and M. Michalak, “Classification performance assessment for imbalanced multiclass data,” Sci. Rep., vol. 14, no. 1, p. 10759, May 2024, doi: 10.1038/s41598-024-61365-z.
H. N. Ngugi, A. A. Akinyelu, and A. E. Ezugwu, “Machine Learning and Deep Learning for Crop Disease Diagnosis: Performance Analysis and Review,” Agronomy, vol. 14, no. 12, p. 3001, Dec. 2024, doi: 10.3390/agronomy14123001.
P. Mittal, “A comprehensive survey of deep learning-based lightweight object detection models for edge devices,” Artif. Intell. Rev., vol. 57, no. 9, p. 242, Aug. 2024, doi: 10.1007/s10462-024-10877-1.
I.-A. Lin, Y.-W. Cheng, and T.-Y. Lee, “Enhancing Smart Agriculture With Lightweight Object Detection: MobileNetv3-YOLOv4 and Adaptive Width Multipliers,” IEEE Sens. J., vol. 24, no. 23, pp. 40017–40028, Dec. 2024, doi: 10.1109/JSEN.2024.3478810.
H. Yu, C. Qian, Z. Chen, J. Chen, and Y. Zhao, “Ripe-Detection: A Lightweight Method for Strawberry Ripeness Detection,” Agronomy, vol. 15, no. 7, p. 1645, Jul. 2025, doi: 10.3390/agronomy15071645.
D. Lu and Y. Wang, “MAR-YOLOv9: A multi-dataset object detection method for agricultural fields based on YOLOv9,” PLoS One, vol. 19, no. 10, p. e0307643, Oct. 2024, doi: 10.1371/journal.pone.0307643.
K. Vinoth and S. P, “Lightweight object detection in low light: Pixel-wise depth refinement and TensorRT optimization,” Results Eng., vol. 23, p. 102510, Sep. 2024, doi: 10.1016/j.rineng.2024.102510.
J. Kerec, A. L. Machidon, and O. M. Machidon, “Deployment-Aware NAS for Lightweight UAV Object Detectors in Precision Agriculture Crop Monitoring,” AgriEngineering, vol. 8, no. 2, p. 43, Feb. 2026, doi: 10.3390/agriengineering8020043.
Y. Liu et al., “FEGW-YOLO: A Feature-Complexity-Guided Lightweight Framework for Real-Time Multi-Crop Detection with Advanced Sensing Integration on Edge Devices,” Sensors, vol. 26, no. 4, p. 1313, Feb. 2026, doi: 10.3390/s26041313.
Z. Ullah, M. Hong, T. Mahmood, and J. Kim, “Systematic Integration of Attention Modules into CNNs for Accurate and Generalizable Medical Image Classification,” Mathematics, vol. 13, no. 22, p. 3728, Nov. 2025, doi: 10.3390/math13223728.
S. Duhan et al., “Investigating attention mechanisms for plant disease identification in challenging environments,” Heliyon, vol. 10, no. 9, p. e29802, May 2024, doi: 10.1016/j.heliyon.2024.e29802.
A. El Hanafy, A. Hessane, and Y. Farhaoui, “Enhancing Deep Learning Models with Attention Mechanisms for Interpretable Detection of Date Palm Diseases and Pests,” Technologies, vol. 13, no. 12, p. 596, Dec. 2025, doi: 10.3390/technologies13120596.
S. Karthikeyan, R. Charan, S. Narayanan, and L. Jani Anbarasi, “Enhanced plant disease classification with attention-based convolutional neural network using squeeze and excitation mechanism,” Front. Artif. Intell., vol. 8, Aug. 2025, doi: 10.3389/frai.2025.1640549.
P. Nasra et al., “Optimized ReXNet variants with spatial pyramid pooling, CoordAttention, and convolutional block attention module for money plant disease detection,” Discov. Sustain., vol. 6, no. 1, p. 391, May 2025, doi: 10.1007/s43621-025-01241-6.
Y. Wang, P. Zhang, and S. Tian, “Tomato leaf disease detection based on attention mechanism and multi-scale feature fusion,” Front. Plant Sci., vol. 15, Apr. 2024, doi: 10.3389/fpls.2024.1382802.
J. Liu, X. Wang, Q. Zhu, and W. Miao, “Tomato brown rot disease detection using improved YOLOv5 with attention mechanism,” Front. Plant Sci., vol. 14, Nov. 2023, doi: 10.3389/fpls.2023.1289464.
E. Yilmaz, S. C. Bocekci, C. Safak, and K. Yildiz, “Advancements in smart agriculture: A systematic literature review on state‐of‐the‐art plant disease detection with computer vision,” IET Comput. Vis., vol. 19, no. 1, Jan. 2025, doi: 10.1049/cvi2.70004.
M. Shoaib et al., “An advanced deep learning models-based plant disease detection: A review of recent research,” Front. Plant Sci., vol. 14, Mar. 2023, doi: 10.3389/fpls.2023.1158933.
S. Duhan, P. Gulia, N. S. Gill, and E. Narwal, “RTR_Lite_MobileNetV2: A lightweight and efficient model for plant disease detection and classification,” Curr. Plant Biol., vol. 42, p. 100459, Jun. 2025, doi: 10.1016/j.cpb.2025.100459.
Y. Ibrahim, M. O. Momoh, K. O. Shobowale, Z. Mukhtar Abubakar, and B. Yahaya, “EDANet: A Novel Architecture Combining Depthwise Separable Convolutions and Hybrid Attention for Efficient Tomato Disease Recognition,” J. Comput. Theor. Appl., vol. 3, no. 2, pp. 160–170, Oct. 2025, doi: 10.62411/jcta.14620.
F. M. Firnando, D. R. I. M. Setiadi, A. R. Muslikh, and S. W. Iriananda, “Analyzing InceptionV3 and InceptionResNetV2 with Data Augmentation for Rice Leaf Disease Classification,” J. Futur. Artif. Intell. Technol., vol. 1, no. 1, pp. 1–11, May 2024, doi: 10.62411/faith.2024-4.
X. Chi, M. Wang, Y. Gao, and Z. Ge, “Deep learning-based assessment of periapical radiographic image quality,” Sci. Rep., vol. 16, no. 1, p. 5047, Jan. 2026, doi: 10.1038/s41598-026-35100-9.
R. N. Asif et al., “Brain tumor detection empowered with ensemble deep learning approaches from MRI scan images,” Sci. Rep., vol. 15, no. 1, p. 15002, Apr. 2025, doi: 10.1038/s41598-025-99576-7.
A. Raza, F. Hanif, and H. A. Mohammed, “Clinical validation of lightweight CNN architectures for reliable multi-class classification of lung cancer using histopathological imaging techniques,” Sci. Rep., vol. 16, no. 1, p. 6512, Jan. 2026, doi: 10.1038/s41598-026-36652-6.
Y. Tian, Q. Ye, and D. Doermann, “YOLOv12: Attention-Centric Real-Time Object Detectors,” ArXiv. Feb. 18, 2025. [Online]. Available: http://arxiv.org/abs/2502.12524
R. Hakani and A. Rawat, “Edge Computing-Driven Real-Time Drone Detection Using YOLOv9 and NVIDIA Jetson Nano,” Drones, vol. 8, no. 11, p. 680, Nov. 2024, doi: 10.3390/drones8110680.
Ultralytics, “YOLOv9: A Leap Forward in Object Detection Technology,” Ultralytics Docs, 2024. https://docs.ultralytics.com/models/yolov9/ (accessed Dec. 22, 2025).
S. Woo, J. Park, J.-Y. Lee, and I. S. Kweon, “CBAM: Convolutional Block Attention Module,” 2018, pp. 3–19. doi: 10.1007/978-3-030-01234-2_1.
L. T. Ramos, E. Casas, C. Romero, F. Rivas-Echeverría, and E. Bendek, “A study of YOLO architectures for wildfire and smoke detection in ground and aerial imagery,” Results Eng., vol. 26, p. 104869, Jun. 2025, doi: 10.1016/j.rineng.2025.104869.
A. Aldubaikhi and S. Patel, “Advancements in Small-Object Detection (2023–2025): Approaches, Datasets, Benchmarks, Applications, and Practical Guidance,” Appl. Sci., vol. 15, no. 22, p. 11882, Nov. 2025, doi: 10.3390/app152211882.
H. T. Van, G. Van Vu, T. Thanh Tuan, B. Vo, and Y. S. Chung, “LGENetB4CA: A novel deep learning approach for chili germplasm Differentiation and leaf disease classification,” Comput. Electron. Agric., vol. 233, p. 110149, Jun. 2025, doi: 10.1016/j.compag.2025.110149.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Miwan Kurniawan Hidayat, Jufriadif Na'am, Ferda Ernawan
This work is licensed under a Creative Commons Attribution 4.0 International License.
