Early Detection of BSR Disease in Oil Palm Trees through Hyperspectral Analysis with MLP-Based Algorithm | Chapter 8 | Advances and Challenges in Science and Technology Vol. 5

 Basal Stem Rot (BSR) affliction, caused by Ganoderma boninense, poses a important threat to Malaysia's oil palm manufacturing, resulting weak yield losses. Detecting BSR effectively is important for maintaining fixed palm oil result. The current method, dependent on manual visual inspection by knowing personnel, demonstrates time-consuming. The combination of depress aerial automobiles (UAVs) and machine learning offers a more effective solution. This study introduces a novel approach to mechanize BSR detection utilizing UAV imagery, enhancing opportunity efficiency and the discovery process. The proposed method includes two key stages: pre-processing hyperspectral concept (HSI) and employing an artificial interconnected system for disease discovery. The Multilayer-Perceptron (MLP) model is introduced to determine spectral features across miscellaneous infection stages. The model is prepared using ground truth dossier collected by prepared surveyors. The HSI dataset includes samples from 2 healthy seedlings, 5 Stage A (mild contamination), 5 Stage B (moderate infection), and 3 Stage C (severe contamination). Performance evaluation involves support vector machines (SVM), 1D convolutional networks (1D CNN), and multiple plants indices in the way that Normalized Difference Vegetation Index (NDVI), Normalized Difference Red Edge (NDRE), Optimised Soil-Adjusted Vegetation Index (OSAVI), and Merris Terrestrial Chlorophyll Index (MTCI). The MLP model exhibits the highest veracity at 86.67%, outperforming SVM and 1D CNN (66.67% and 73.33%, respectively). While plants indices generally detect Stage C trees and struggle to change between Healthy, Stage A, and Stage B wood, the MLP model offers a balanced performance accompanying moderate training period and quicker inference occasion. This demonstrates the model's influence in detecting BSR, even at an early infection stage.

Author(s) Details:

Chee Cheong Lee,
Faculty of Engineering and Technology, Multimedia University, 75450 Bukit Beruang, Melaka, Malaysia.

Voon Chet Koo,
Faculty of Engineering and Technology, Multimedia University, 75450 Bukit Beruang, Melaka, Malaysia.

Tien Sze Lim,
Faculty of Engineering and Technology, Multimedia University, 75450 Bukit Beruang, Melaka, Malaysia.

Yang Ping Lee,
FGV R&D Sdn. Bhd. 50350 Kuala Lumpur, Malaysia.

Haryati Abidin,
FGV R&D Sdn. Bhd. 50350 Kuala Lumpur, Malaysia.

Please see the link here: https://stm.bookpi.org/ACST-V5/article/view/12272