A novel dataset and deep learning object detection benchmark for grapevine pest surveillance

Wine aroma represents one of the most important quality parameter and it is influenced by various factors (viticulture and vinification techniques, climate or storage conditions etc.). Wines produced from conventionally and ecologically grown grapes of same variety have different chemical composition and aroma profile [1]. Aroma profile of wine can be also influenced by additional treatment of wine, such as concentration of wine by reverse osmosis (RO). Reverse osmosis represents a pressure-driven membrane separation technique that separates the initial wine on the retentate or concentrate that is retained on the membrane, and permeate that passes through it [2]. Wine permeate usually containes water, ethanol, acetic acid and several low molecular weight compounds that can pass through the membrane. This property enables the use of reverse osmosis membranes for wine concentration, partial dealcoholization, acetic acid or aroma correction [3,4].

Increasing concentrations of greenhouse gases (GHGs) in the atmosphere due to human activities are leading to a rise in the average temperature of the atmosphere. among the scenarios established by the un’s intergovernmental panel on climate change (IPCC), only two enable us to achieve the minimum objective of the paris agreements signed at cop 21 in 2015: staying below +2°c after 2050. both scenarios forecast a rapid reduction in GHG emissions as early as 2025, thanks to strong international cooperation, the priority given to sustainable development and responsible consumer choices.

Geospatial technologies have significant contribution to viticulture, especially in small-scale vineyards, which require precise management. Geospatial data collected by modern technologies, such as Unmanned Aerial Vehicle (UAV) and satellite imagery, can be processed by modern software and easily be stored and transferred to GIS environments, highlighting important information about the health of vine plants, the yield of grapes and the wine, especially in wine-making varieties. The identification of field variability is very important, particularly for the production of high quality wine. Modern geospatial data management technologies are used to achieve an easy and effortless localization of the fields’ variability.

We propose here a brand new large routine microbiological analysis method intended for oenology, in flow cytometry, using high performance equipment and triple selective cell staining, activated by fluorescence. The results and practical applications of the method are presented: Brettanomyces (Dekkera) Monitoring, fermentations monitoring, bottling and enological practices monitoring.The method allow a complete new microbiological tool for wine industry.The method has been accredited ISO 17025 in our laboratories.

Deficit irrigation strategies can be valuable means to improve grape quality while saving important amounts of water. A simple way to use deficit irrigation can be based on irrigating a vineyard with a determined level of crop evapotranspiration. Using a precise physiological parameter indicating water status, irrigation could be managed to maintain a specific pre-dawn leaf water potential.