Electronic nose for early detection of basal stem rot caused by Ganoderma in oil palm

Abstract

The successful control of basal rot disease (BSR) determined by early detection of infection because when the symptoms already appear, generally plants are difficult to save. The earlier the Ganoderma infection is known, the easier the control will be and the losses can be minimized. Therefore, early detection of Ganoderma infection is very necessary, which in this study was carried out by detecting volatile compounds using electronic nose (E-nose). E-nose detection has been carried out to analyze the compounds formed in pure Ganoderma culture. Detection of plants in the field carried out at 4 levels of infection, i.e. healthy, early, moderate and severe infection. The results concluded that Ganoderma mycelium when compared with other fungi (Trichoderma, Aspergillus and Omphalina) showed significant differences when analyzed using an unsupervised PCA chemometric system. The E-nose data processed using machine learning Support Vector Machine (SVM) was able to distinguish the aroma between Ganoderma boninense CSB, G. boninense ‘Rejosari’, and G. lucidum with an accuracy rate of 99.64%. E nose was able to differentiate with high accuracy (90.95%) of each infection level even though there was still a slice between in root sample.