Customized YOLO26 Architecture Improves Grapevine Disease Detection on Embedded Devices

Written by

in

Researchers developed the YOLO-Gr framework based on an improved YOLO26 model that can run directly on low-power embedded hardware to identify grapevine leaf diseases. It achieved 97.9% precision, 96.1% recall, and 98.4% mean average precision at IoU 0.50. By integrating mobile, edge, and cloud architectures, it supports rapid on-site screening in vineyards.

Source:BIOENGINEER.ORG · Read original · 前沿热点

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *