Research Article
Cotton Disease Detection on UAV Images: A Deep Learning-Based Approach with YOLOv7
@INPROCEEDINGS{10.1007/978-3-031-81573-7_19, author={Zakaria Kinda and Sadouanouan Malo and Thierry Roger Bayala and Issa Wonni}, title={Cotton Disease Detection on UAV Images: A Deep Learning-Based Approach with YOLOv7}, proceedings={Towards new e-Infrastructure and e-Services for Developing Countries. 15th International Conference, AFRICOMM 2023, Bobo-Dioulasso, Burkina Faso, November 23--25, 2023, Proceedings, Part II}, proceedings_a={AFRICOMM PART 2}, year={2025}, month={2}, keywords={Deep Learning UAV images Cotton diseases YOLOv7}, doi={10.1007/978-3-031-81573-7_19} }- Zakaria Kinda
Sadouanouan Malo
Thierry Roger Bayala
Issa Wonni
Year: 2025
Cotton Disease Detection on UAV Images: A Deep Learning-Based Approach with YOLOv7
AFRICOMM PART 2
Springer
DOI: 10.1007/978-3-031-81573-7_19
Abstract
Cotton is the most important agricultural product in Burkina Faso, and it is farmed by 25% of the country's population. Cotton diseases, on the other hand, are a big issue for this crop, contributing considerably to output losses. These disorders are detected manually, which increases the time required for therapy. Artificial intelligence techniques can help enhance cotton production by automatically recognizing these diseases. This research aims to detect cotton diseases using UAV photos obtained in the field. We used the YOLOv7 detection model fine-tuned on the tomato leaves dataset and subsequently applied to the cotton leaves dataset. On the cotton dataset, experimental results from the YOLOv7 model yielded mAP@0.50 (mean average precision), f1score, Precision, and Recall of 50.7%, 55.4%, 53.7%, and 57.4%, respectively.
