Citation

Khoo, Brian Kai Yuan and Lew, Sook Ling and Avtar, Ram (2024) Oil palm tree health status detection using canopy height model and NDVI. In: MULTIMEDIA UNIVERSITY ENGINEERING CONFERENCE 2023 (MECON2023), 26–28 July 2023, Virtual Conference.

Full text not available from this repository.

Abstract

Oil palm plantations span across extensive land areas, making it challenging for plantation owners to individually monitor the health of each tree. It is very time consuming and requires large amount of manpower to evaluate each tree’s health and locate the trees that has problem. The plantation’s owner might consider replacing the dead trees and curing the unhealthy trees. Therefore, the owner requires a mechanism to swiftly assess the health status of the oil palm trees and pinpoint their locations, enabling prompt action to mitigate losses. In this research, the objectives are using vegetation indices to analyze the health status into three distinct categories of oil palm trees. The method to detect the trees in the UAV image is local maxima of Canopy Height Model (CHM) and the health status of oil palm trees is assessed by employing Normalized Difference Vegetation Index (NDVI). It is concluded that the proposed method using CHM perfonned well in detection of oil palm trees with 83.10% accuracy. Moreover, classifying health by employing NDVI stands at 94.90% accuracy. Hence, this study has effectively addressed the challenge of detecting oil palm trees in low-resolution UAV images through the utilization of local maxima within the CHM. This study also has proven that NDVI is suitable to analyze the oil palm trees health and introduce a modeling approach to further enhance the accuracy and predictive capabilities of NDVI.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Remote sensing
Subjects:	G Geography. Anthropology. Recreation > G Geography > G1-922 Geography (General)
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	04 Dec 2024 02:47
Last Modified:	04 Dec 2024 02:47
URII:	http://shdl.mmu.edu.my/id/eprint/13211

Downloads

Edit (login required)