Disease Detection in Grape Cultivation Based on Deep Learning Algorithms

Vargas J.P.B.L.; Pérez Paredes M.G.S.; Ruelas A.M.R.; Barrios R.J.C.; de Paula M.V. · Lecture Notes in Networks and Systems (2026) · DOI: 10.1007/978-3-032-00236-5_11

Resumen

Recent advancements in deep learning have significantly enhanced image recognition capabilities, presenting substantial opportunities for applications in agriculture, such as disease detection. Grapevines flourish in Peru’s warm tropical regions but are vulnerable to climate fluctuations and pest infestations, posing risks to crop yields and export value. This study proposes an image-processing approach to detect three common diseases affecting grapevine leaves. We leveraged deep learning algorithms, including ResNet, MobileNet, EfficientNet, AlexNet, GoogLeNet, DenseNet, and YOLO-NAS, the last one within the Roboflow framework across various image capture scenarios. Experimental results indicate that YOLO-NAS outperforms the others, achieving 98.4% accuracy, a mean average precision of 99.9%, and a recall of 100%. These promising findings highlight the potential of deep learning technologies to support decision-making in grapevine management, improving the efficiency of pesticide use, fertilizer application, irrigation adjustments, and crop protection strategies. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.

Instituciones: Universidad Nacional de Ingeniería; Universidad Tecnológica del Perú