Diego Freire Bastidas, Alessandro De Rosa, and Luca Brillante*
*Department of Viticulture and Enology, California State University Fresno, 2360 E Barstow Ave, Fresno, CA, 93740 (lucabrillante@csufresno.edu)

The development of new varieties has played a significant role in maintaining a strong and stable food supply. New plant varieties enable consumers to enjoy diverse, safe, nutritious, and abundant food and are protected under the Plant Variety Protection Act. The protection of intellectual property encourages continued investment in research and development and encourages bringing innovations to farmers to help them cope with changing conditions, from the climate to the markets. Protected plants can still be vegetatively propagated illegally. This undermines investment in research and development of new varieties and sustainable agricultural technologies.

This study aimed to investigate the ability of machine learning to identify grapevine varieties in Sentinel 2 satellite imagery. Our preliminary results, employing dimension reduction techniques alongside rigorous cross-validation, show an accuracy rate of >84% on >400 accessions within a single location. This accuracy underscores the efficacy of our approach in effectively discriminating among a diverse array of grapevine varieties present within the ranch.

Funding Support: N/A