MONITORING THE GROWTH AND QUALITY OF JAPANESE GREEN TEA BY UAV IMAGERY

Abstract

The optimal harvest season of Japanese green tea in Japan is judged by individual
farmers based on their experiences. To confirm that optimal season, it requires a lot
of efforts such as sampling of tea flushes by plucking, chemical component
measuring by near-infrared spectroscopic analysis. Thus, an efficient method to
determine the optimal season which can cover a wide area at once is considered
highly helpful for farmers. This research measured normalized difference
vegetation index (NDVI) of tea flushes by a multi-spectrum sensor mounted on an
unmanned aerial vehicle. At the same time, chemical component of tea flushes was
measured to compare with the result of NDVI. The result showed that the value of
NDVI for the optimal harvest season of tea-leaf is 0.6 to 0.7. The result of chemical
component measuring showed the increase of total free amino acids and reduction
of total free sugars while tea-leaf ripens. Therefore, there was a positive relation
between the value of NDVI, and total free amino acids and total free sugars. It
indicates a high possibility to predict the optimal harvest season of tea flushes by
NDVI.