UPM Institutional Repository

Severity of Ganoderma boninense disease classification using SAR data


Che Hashim, Izrahayu and Mohamed Shariff, Abdul Rashid and Bejo, Siti Khairunniza and Muharam, Farrah Melissa and Ahmad, Khairulmazmi (2018) Severity of Ganoderma boninense disease classification using SAR data. In: 39th Asian Conference on Remote Sensing (ACRS 2018), 15-19 Oct. 2018, Renaissance Kuala Lumpur Hotel, Malaysia. (pp. 2492-2499).


Basal stem rot disease (BSR) in oil palm plantations is caused by Ganoderma boninense fungus. BSR is a major disease attacking oil palm plantations in Malaysia and Indonesia. But for now, the only available treatment is to prolong the life of oil palm trees as there is no effective treatment for BSR. To control this disease, early detection of G. Boninense infection is a decent strategy. Many researchers have used remote sensing techniques for early detection and mapping of BSR disease in oil palm plantations based on G. boninense infection symptoms. The main objective of this project is to study the potential of radar backscattering for classifying severity level of G. boninense disease in oil palm plantation. The processing stage involved the usage of two different machine learning algorithms to classify severity level of G. boninense in oil palm plantation. This study uses Alos Palsar 2 image with dual polarization; HH (Horizontal - transmit and Horizontal - receive) and HV (Horizontal - transmit and Vertical - receive) archived on March 20, 2017. Two classifier models: Multilayer Perceptron (MP) and Kstar are tested by using Weka open source software. The MP classifier model for HV polarization is the best for predicting and classifying severity level of G. boninense in oil palm plantation in terms of correctly classified. Model MP classifier and HV polarization reach 77.17% correctly classified. In addition, this study can separate oil palm by severity of each T0 (92.73%), T1 (0%), T2 (93.33%) and T3 (54.55%).

Download File

[img] Text
39TH ACRS-6.pdf
Restricted to Repository staff only

Download (497kB)

Additional Metadata

Item Type: Conference or Workshop Item (Paper)
Divisions: Faculty of Agriculture
Faculty of Engineering
Keywords: Basal stem rot; Radar backscattering; Machine learning; Classifier
Depositing User: Nabilah Mustapa
Date Deposited: 06 Mar 2019 05:38
Last Modified: 06 Mar 2019 05:38
URI: http://psasir.upm.edu.my/id/eprint/67019
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item