Study of the oil palm crown characteristics associated with Basal Stem Rot (BSR) disease using stratification method of point cloud data

•3D image of oil palm trees were extracted using TLS technology.•Crown profiles were developed to identify the healthy and unhealthy trees.•Crown strata were used to differentiate the healthy and unhealthy trees.•Proposed method can classify healthy and unhealthy trees with 92.5% accuracy. Basal Stem Rot (BSR) disease caused by the Ganoderma boninense fungus is the most serious issue affecting the oil palm industry. Terrestrial laser scanning (TLS) technology is a precise, contact-free, and active remote sensing method which is widely used for determining tree morphology using the crown strata. However, its capability in BSR detection using crown strata solely has not been explored. Therefore, this study used TLS data to study the effect of crown profile due to BSR and develop a BSR detection model using crown strata. Ninety-two palms were selected at random consists of 36 healthy oil palm trees and 56 unhealthy oil palm trees infected by Ganoderma boninense. The palms were each scanned at four different positions around them, and crown strata (named as Cn; where n is the crown length in cm) were generated using crown stratification method. The numbers of laser hits in the strata were used to generate crown profiles comparing healthy and unhealthy oil palm trees, to develop prediction models and to examine the patterns of crown density in different parts. The study has accounted for the occlusion problem by carrying out a systematic multi-scan approach. Moreover, the short traveling distance (less than 12 m) needed for the laser to reach the maximum crown tips ensures greater point density. The profiles of crown strata revealed that the healthy trees have higher crown densities than unhealthy trees starting from 250 cm from the top (strata no. 5) to the bottom. Prediction models using the strata parameters C650, C700, C800, and C850 were 92.5% accurate in the classification of healthy-unhealthy trees.