HEIGHT MEASUREMENT AND OIL PALM YIELD PREDICTION USING UNMANNED AERIAL VEHICLE (UAV) DATA TO CREATE CANOPY HEIGHT MODEL (CHM)

Abstract

Oil palms are currently in high demand, which tend to increase even higher as a source of alternative energy for humans, especially in Southeast Asian countries. This leads to the study that focuses on the height measurement, using an unmanned aerial vehicle (UAV), and age analysis of oil palm trees planted within the experimental plots in order to predict their yield. The methodology described in the paper provides using Canopy Height Model (CHM) for height measurement and prediction of the oil palm yield by multiple linear regression. The results indicated that the errors caused by overlapping age ranges were found in 3 out of 12 experimental plots. Furthermore, the primary factors influencing the oil palm yield prediction included the age (9 years and above) and canopy density (over 41% of the area), while the secondary factors supporting more accuracy included the total plot area, canopy area, and canopy height, with the coefficient of determination or R-squared at 0.98. In this study, we learned that the aforementioned factors could be concluded from the data collected by an UAV, which reduced the time for measuring the height of each tree manually, resulting in more accurate yield prediction.