Deep Learning in Agriculture: Detection and Analysis of Sugar Beets with YOLOv8

ADBA Computer Science is a peer-reviewed interdisciplinary journal dedicated to advancing understanding and application of all areas within computer science. The journal serves as a nexus for researchers, educators, engineers, and practitioners to disseminate cutting-edge research findings, innovative theories, and practical applications across diverse computer science disciplines. The journal aims to foster interdisciplinary collaboration and facilitate the exchange of ideas among researchers and practitioners working in various fields of computer science. We welcome contributions that explore theoretical developments, computational methods, experimental observations, and practical implementations related to computer science.

Öz

In this study, the performance of the YOLOv8 model in detecting sugar beets was evaluated usingimages obtained from a drone over a sugar beet field. High-resolution drone images were divided into smallsegments, labeled, and the model was trained using data augmentation techniques. The results obtainedduring the training and testing phases demonstrated that the model successfully detected sugar beets withhigh accuracy, precision, recall, and F1 score values. The analysis of label correlograms and result graphsconfirmed the model’s labeling accuracy and detection capability. These findings indicate that the YOLOv8model can be an effective tool in agricultural production monitoring and plant health assessment applications.In the future, the model’s performance will be more comprehensively evaluated using datasets obtained fromdifferent geographical regions and various agricultural products.