DISEASE DETECTION FROM FIELD SPECTROMETER DATA

Oil palm plants have been planted in a large scale of areas. However, Ganoderma disease has been recognized and diagnosed in the oil palm plants that has infected almost a half of the oil palm plants in Malaysia. To deal with this problem, the use of vegetation indices analysis on hyper spectral field data, this paper examines the ability of spectral data in identifying the stages of Ganoderma disease. The favourable result will be helpful to control the spreading of the diseases. By using vegetation indices, the oil palm plants could be classified into three categories, namely: 1 (T1 healthy), 2 (T2 semi healthy) and 3 (T3 severe damage). The results show that the best vegetation index is the Modified Red Edge Simple Ratio (MSR705) among the vegetation indices to identify the oil palm health stages. Moreover, it has been observed that the index of Narrowband greenness VIs has been exhibited an acceptable outcome in differentiating between the oil palm plant stage 1 (T1 healthy) and stage 2 (T2 semi healthy).