<![CDATA[Whitefly (Aleyrodidae) pests pose a significant threat to the productivity of Cucurbitaceae crops in Indonesia, leading to substantial harvest losses and the transmission of plant viruses. Because traditional manual detection methods are often slow and inefficient, there is a clear need for a technological solution for early identification. This study details the development and evaluation of an object detection model using the YOLOv11 architecture. The methodology involved four primary stages: preparing a dataset of 1,940 images from public repositories, preprocessing the data through annotation and augmentation (including blur, brightness, and noise), training the model, and conducting a thorough performance evaluation. The resulting model was deployed into a web-based application for real-time detection. The evaluation demonstrated the model's excellent performance, achieving a mean Average Precision at a 0.5 IoU threshold (mAP@50) of 85.6% and an mAP@50-95 of 81.2%. Furthermore, it achieved a precision of 83.1%, a recall of 89.0%, and an F1-Score of 86.0%, proving its capacity to consistently and accurately detect these small-sized pests. This research successfully delivers an effective and accessible early detection system, making a practical contribution to precision agriculture and supporting food security in Indonesia through the application of deep learning.]]>
Copyrights © 2025
