AI New Zealand

View Original

Grape Expectations: Putting AI to Work in the Vineyard.

Lincoln Agritech, a research and development company owned by Lincoln University, is developing an AI solution which can make early-season predictions of vineyard harvests.

“Grape growers and wineries spend a lot of money trying to predict their grape yield each year,” says Lincoln Agritech optics and image processing team leader Jaco Fourie.

“This currently involves hiring a large number of workers to manually sample grape bunches.”

Through a project funded by the Ministry of Business, Innovation and Employment and NZ Winegrowers, Lincoln Agritech is working on creating a system that instead uses electronic sensors to accurately count grapes.

“The sensors will capture and analyse grape bunches within individual rows, and assess the number, sizes and distribution,” says Dr Fourie.

“We’ll then feed this data into computer algorithms, which have been designed by the University of Canterbury, to predict grape yield at harvest time.”

New data will be added to the system each year, leading to continuous improvements in the model’s accuracy as more information is gathered under different conditions.

Lincoln Agritech CEO Peter Barrowclough says “the game-changing innovation will enable growers to accurately assess differences in yield not only between regions or vineyards but also blocks and rows.”

“Over the long term, site-specific yield prediction will help reduce costs by enabling better planning both in the vineyard and in market.”

NZ Winegrowers’ general manager of research and innovation,
Dr Simon Hooker, says the technology will benefit the industry by supporting better crop management, smoother processing and market forecasting based on capacity to supply.

Collaborating partners on the project include Plant and Food Research, Lincoln University, the University of Canterbury, CSIRO (Adelaide), NZ Winegrowers and local winegrowers in the Marlborough region.

// Source - https://www.callaghaninnovation.govt.nz/sites/all/files/ai-whitepaper.pdf