| Convolutional SVM Networks for Early Detection of Ganoderma boninense in Oil Palm Using UAV and Multispectral Pleiades Images |
| Paper ID : 1179-GEOSPATIAL (R6) |
| Authors: |
|
parisa Ahmadi *1, shattri Mansor2, hamidreza Ahmadzadeh Araji3, Bil lu4 1Geography Department Simon Fraser University, 8888 University Dr., Burnaby, B.C, Canada 2Department of Civil Engineering, Faculty of Engineering, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia 3Texas A&M AgriLife Research& Extension Centre,1509 Aggie Drive,Beaumont, Tx 77713,USA 4Department of Geography, Simon Fraser University, 8888 University Drive, Burnaby, BC V5A 1S6, Canada |
| Abstract: |
| Oil palm plays a significant role in Malaysia’s economy as Malaysia is the second-largest palm oil producer in the world. In oil palm plantations, Ganoderma boninense is a causal agent of the basal stem rot (BSR) disease that is responsible for a considerable annually losses, especially in South East Asia. The disease remains an unresolved problem in most production areas due to lack of disease management strategy to detect the infected palms at their early stage. In recent years, advancement in remote sensing platforms and image processing methods have produced remarkable results for the early detection of diseases. In this study, support vector machine (SVM) classifier was performed on UAV and Pleiades imagery to identify the best classification model for early detection of BSR disease in oil palms. The results of this study suggested that the best prediction result was obtained from UAV with the overall accuracy of 68.28% while 64.52% for the Pleiades, whereas the early Ganoderma infection could be detected with an accuracy of 64.07% and 64.49%, respectively. |
| Keywords: |
| Keywords: Basal stem rot; Support Vector Machine; UAV; Pleiades |
| Status : Paper Accepted (Poster Presentation) |
