Researchers at Germany's University of Bonn and Julius Kuhn-Institut have developed a mobile laboratory to automate and accelerate the process of checking the wax coverage of grapes, a labor-intensive task that is prone to human error.
The system works by taking a series of high-resolution photos of the grapes in order to pick up patterns of scattered light, including the diffuse reflections emitted by the roughness of the wax. The images were analyzed by two artificial intelligence (AI) algorithms: one singled out the grape in the image, while another determined the distribution of waxiness over the grapes.
The team trained the AI algorithms on 90 images with more than 6.5 million labelled pixels, and then took 180 images of six different grape varieties to train and test the system.
The team found the system could detect wax distribution on a grape with an accuracy of about 95%.
From New Scientist
View Full Article
Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA
No entries found