Host: The Japanese Society for Artificial Intelligence
Name : 34th Annual Conference, 2020
Number : 34
Location : Online
Date : June 09, 2020 - June 12, 2020
The most important parameter for consumers choosing vegetables is freshness. Cumulative temperature (CT) after harvest, a product of storage temperature and time, has been proposed as an objective freshness indicator. However, CT is hard to be monitored throughout the whole vegetable distribution process. In this study, the potential of visible-near infrared (VIS-NIR) spectroscopy for non-destructive CT estimation of vegetables was investigated. Freshly harvested komatsuna leaves were stored at 5, 10, and 20 degrees Celsius. Their VIS-NIR absorption spectra were measured daily in the wavelength range of 400-2500 nm. Then samples were freeze-dried for proton nuclear magnetic resonance (NMR) spectroscopy metabolomics analysis. CT could be predicted with the coefficient of determination of prediction of 0.75 by partial least squares regression of VIS-NIR spectra. A correlation analysis between VIS-NIR and 1H NMR spectra indicated that CT could be predicted because changes in sugars and acids were reflected to VIS-NIR spectral changes.