Using mobile and drone-based cameras to create inventories and documentation for vineyards

The impacts of climate change and increasingly extreme weather conditions are posing significant challenges for winegrowers in their daily operations. Determining the best management strategies to minimize yield loss, reduce variability, and avoid unnecessary plant protection measures has become more crucial than ever. One valuable resource is the use of image-based field documentation and disease monitoring. In the COMMECT project, we have developed multi-scale methods to generate such data, enabling improved management at the sub-field level. For example, monitoring Esca—a widespread trunk disease—allows for targeted interventions during the winter season, even when symptoms are no longer visible. By collecting tractor- and drone-based images and analyzing them with machine-learning techniques, we provide reliable, actionable information to support winegrowers in managing their vineyards more effectively.

Image trap data for pest monitoring

The most important pest in olive groves, the olive fly, causes significant yield and quality losses. In the fight against this insect, either a standard spraying schedule is used, neighboring practices are followed, or no spraying is done at all. In these cases, environmental effects increase with unnecessary or untimely spraying, and the fight against the insect is unsuccessful. In the COMMECT project, olive fly tracking can be done remotely with a system that can send data with the help of cameras and sim cards integrated into yellow sticky traps that are made attractive with the help of pheromones. This system, which we call digital traps, receives its energy with the help of solar panels and uses its battery. The most important benefit of the system is that it provides data for spraying at the right time and in the right amount. On the other hand, the fact that the pest can be diagnosed without the need for physical access to the olive groves, which are mostly in difficult terrain conditions, makes things even easier. The next step is to recognize the olive fly by artificial intelligence and make the images usable by the producer without the need for an expert opinion.

SPEAKERS

Miriam Machwitz is a senior researcher at LIST, holding a PhD in remote sensing and vegetation modeling. Her research focuses on detecting diseases and crop stress in viticulture and field crops using remote sensing data across various spectral domains, from drone to satellite scales.

Mario Gilcher is coordinator of the GEODATA development team of LSC360. Building on his doctoral research in machine learning and statistical modeling of earth observation data, he now develops and streamlines photogrammetry pipelines and AI-powered mapping solutions.

Mehmet Hakan is a senior researcher at Olive research Institute. He started his agriculture education life at Beydere Agriculture High School and completed his PhD at Ege University Horticulture Department. He took part in many projects in the Breeding and Genetics department of Izmir Olive Research Institute, where he was appointed in 2008, and he still continues to work in the same department.