VineKeeper

Intelligent Identification and Digital Curation of Portuguese Wine Heritage

 

Context:

In Portugal, particularly in the Douro and Alentejo regions, there is a unique wine-growing heritage that faces threats such as the homogenization of plantations, climate change, and the spread of diseases. At the same time, winegrowers need accessible tools to accurately identify grape varieties and diagnose diseases at an early stage, reducing the use of phytopharmaceuticals and increasing the sustainability of production. The VineKeeper project arises from the need to preserve and enhance Portugal’s wine-growing heritage, using advanced technologies to identify, catalogue, and preserve the country’s indigenous grape varieties, with special attention to minority varieties and, in some cases, those at risk of extinction.

 

Goals:

VineKeeper is an innovative digital platform that turns mobile devices into advanced tools for identifying, cataloguing, and protecting Portugal’s indigenous grape varieties, democratizing access to them for winegrowers. Focusing mainly on technologies based on Computer Vision (CV) and Artificial Intelligence (AI) for varietal recognition and disease identification, the system aims to promote the preservation of Portugal’s valuable wine heritage. In particular, an innovative AI-based integrated software solution will be implemented, comprising a mobile App and a Web App. The mobile app will support the user in capturing images in the field, namely through the automatic detection of key parts of the vineyard for the target tasks, and the respective automatic validation of image quality. The Web App will also include CV and AI modules for the automatic identification of grape varieties and the automatic identification of diseases that limit the vegetative multiplication of vines.

 

Impact:

The results of the VineKeeper project will be integrated and commercialized as a new optional module for the existing agricultural management platform, SIGP – Sistema Integrado de Gestão de Propriedades, allowing GeoDouro to expand its commercial offer. The solution will be validated in a real environment, in the productive context of current or potential SIGP clients. Real data will be used to develop and train the AI algorithms for automatic image processing, to achieve a minimum TRL of level 6 and demonstration potential at TRL 7.

 

Consult the project's Spec Sheet here:

English

 

Funded by:

For any additional information, please contact us using the inquiries form.