The estimation of nutrient content using multispectral image analysis in palm oil ( Elaeis guineensis Jacq)

Abstract

Indonesia is the largest producer and exporter of palm oil in the world. 50% -70% of operational costs and around 25% of total production costs in oil palm plantations are only for fertilizer. The requirements for effective fertilizer management are meeting plant nutrition requirements and preventing nutrient deficiencies. This study aims to estimate the nitrogen, phosphorus and potassium nutrients contained in palm oil trees using multispectral cameras taken with drones. The method used was divided into data preparation and leaf sampling and taking pictures with drones. Second, pre-data processing, the things that are done was photo stitching, georeferencing and digitizing the land and oil palm canopy. Third, the data analysis stage found the correlation between nutrition and image analysis using regression analysis with the backward method. The results of this study are that the nitrogen model has correctness of 95.11%, the phosphorus model has correctness of 95.38%, and the potassium model has correctness of 88.65%.