TY  - JOUR
T1  - Towards intelligent pruning of vineyards by direct detection of cutting areas
A1  - Pacioni, Elia
A1  - Abengózar, Eugenio
A1  - Macías Macías, Miguel
A1  - García Orellana, Carlos J
A1  - Gallardo, Ramón
A1  - González Velasco, Horacio M
JA  - MDPI Agriculture, Digital Agriculture
Y1  - 2025
VL  - 15
IS  - 11
UR  - https://www.mdpi.com/2077-0472/15/11/1154
KW  - computer vision
KW  - Mask R-CNN
KW  - Object segmentation
KW  - Vineyard pruning
KW  - YOLOv8
N2  - The development of robots for automatic pruning of vineyards using deep learning techniques seems feasible in the medium term. In this context, it is essential to propose and study solutions that can be deployed on portable hardware, with artificial intelligence capabilities but reduced computing power. In this paper, we propose a novel approach to vineyard pruning by direct detection of cutting areas in real time by comparing Mask R-CNN and YOLOv8 performances. The studied object segmentation architectures are able to segment the image by locating the trunk, and pruned and not pruned vine shoots. Our study analyzes the performance of both frameworks in terms of segmentation efficiency and inference times on a Jetson AGX Orin GPU. To compare segmentation efficiency we used the mAP50 and AP50 per category metrics. Our results show that YOLOv8 is superior both in segmentation efficiency and inference time. 
Specifically, YOLOv8-S exhibits the best tradeoff between efficiency and inference time, showing an mAP50 of 0.883 and an AP50 of 0.748 for the shoot class, with an inference time of around 55 ms on a Jetson AGX Orin.
ER  -