Food Authenticity Seminar Series: Machine learning-driven hyperspectral imaging for origin traceability of green coffee
Coffee consumption is rising globally, especially for single-origin specialty coffee. This surge in demand has led to more cases of geographical origin fraud, requiring effective tools to verify coffee origins. In a recent study, Dr. Biniam Kebede and his team applied a multi-omics and machine learning approach to analyze 24 green coffee bean samples (Arabica, wet-washed) from 3 continents, 8 countries, and 22 regions. Multi-omics techniques included geochemistry, MS and NMR-based metabolomics, and vibrational spectroscopy. Various (non)linear machine learning models were explored. NIR-coupled hyperspectral imaging showed strong potential due to its ability to analyze large volumes of beans for more accurate origin traceability.
The Virtual Seminar Series on Food Authenticity is organized by researchers and students at 91社区 and the University of Guelph.
The event is virtual and will take place over Zoom.
About the speaker:聽Dr. Biniam Kebede
Dr. Biniam Kebede holds a PhD in Bioscience Engineering from KU Leuven (Belgium), has worked in New Zealand and will be joining the University of Guelph (Canada) as an Assistant Professor on December 1, 2024.
His active research areas include: (i) understanding process-structure-function relations as a basis for (reverse) engineering approaches to create safe and healthier food products with excellent sensory qualities in a sustainable way and (ii) developing innovative, non-invasive sensors and techniques coupled with machine learning and AI, for the agri-food industry.