A Multi-Layer Perceptron Based Approach for Early Detection of BSR Disease in Oil Palm Tree Using Hyperspectral Image

Abstract

Basal Stem Rot (BSR) disease caused by Ganoderma boninense is identified as the biggest threat in oil palm industry in Malaysia, resulting in significant yield losses. Effective BSR disease detection is important for plantation management to ensure stable palm oil production. BSR disease detection is currently done by experience personnel, via visual inspection and GPS marker. This is labour intensive and very time consuming. This paper proposed a new framework to automate BSR disease detection with UAV images to improve time efficiency and automate detection process. The proposed method has two steps, first hyperspectral image pre-processing, followed by artificial neural network disease detection. Multilayer-Perceptron model is introduced to learn spectral features from different infection stages. The model is trained with ground truth collected by trained surveyors and being marked in RGB images. Performance is compared with several vegetation indices (NDVI, NDRE, OSAVI and MTCI). Result shows proposed method can detect BSR disease particularly Stage C. Further tuning on MLP is done, and overall accuracy of proposed method is 86.67% in detecting Stage A, Stage B, Stage C and Healthy plant. This method demonstrates its effectiveness on BSR disease detection, even at early infection stage.