Impact of Phenological and Lighting Conditions on Early Detection of Grapevine Inflorescences and Bunches Using Deep Learning
Reliable early-stage yield forecasts are essential in precision viticulture, enabling timely interventions such as harvest planning, canopy management, and crop load regulation. Since grape yield is directly related to the number and size of bunches, the early detection of inflorescences and bunches...
| Autores: | , , , , , , |
|---|---|
| Formato: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2025 |
| País: | España |
| Recursos: | Consejo Superior de Investigaciones Científicas (CSIC) |
| Repositorio: | DIGITAL.CSIC. Repositorio Institucional del CSIC |
| OAI Identifier: | oai:digital.csic.es:10261/404062 |
| Acesso em linha: | http://hdl.handle.net/10261/404062 https://api.elsevier.com/content/abstract/scopus_id/105011537702 |
| Access Level: | acceso abierto |
| Palavra-chave: | Deep learning Grape bunch Inflorescence Object detection Precision viticulture Yield prediction Yolov11 |
| Resumo: | Reliable early-stage yield forecasts are essential in precision viticulture, enabling timely interventions such as harvest planning, canopy management, and crop load regulation. Since grape yield is directly related to the number and size of bunches, the early detection of inflorescences and bunches, carried out even before flowering, provides a valuable foundation for estimating potential yield far in advance of veraison. Traditional yield prediction methods are labor-intensive, subjective, and often restricted to advanced phenological stages. This study presents a deep learning-based approach for detecting grapevine inflorescences and bunches during early development, assessing how phenological stage and illumination conditions influence detection performance using the YOLOv11 architecture under commercial field conditions. A total of 436 RGB images were collected across two phenological stages (pre-bloom and fruit-set), two lighting conditions (daylight and artificial night-time illumination), and six grapevine cultivars. All images were manually annotated following a consistent protocol, and models were trained using data augmentation to improve generalization. Five models were developed: four specific to each condition and one combining all scenarios. The results show that the fruit-set stage under daylight provided the best performance (F1 = 0.77, R<sup>2</sup> = 0.97), while for inflorescences, night-time imaging yielded the most accurate results (F1 = 0.71, R<sup>2</sup> = 0.76), confirming the benefits of artificial lighting in early stages. These findings define optimal scenarios for early-stage organ detection and support the integration of automated detection models into vineyard management systems. Future work will address scalability and robustness under diverse conditions. |
|---|