Although hyperspectral remote sensing has been used to study many agricultural phenomena such as crop stress and diseases, the potential use of this technique for detecting Ganoderma disease infestations and damage to oil palms under field conditions has not been explored to date. This research was conducted to investigate the feasibility of using a portable hyperspectral remote-sensing instrument to identify spectral differences between oil-palm leaves with and without Ganoderma infections. Reflectance spectra of samples representative of three classes of disease severity were collected. The most significant bands for spectral discrimination were selected from reflectance spectra and first derivatives of reflectance spectra. The significant wavelengths were identified using one-way analysis of variance. Then, a Jeffries–Matusita (JM) distance measurement was used to determine spectral separability between the classes. A maximum likelihood classifier method was used to classify the three classes based on the most significant wavelength spectral responses, and an error matrix was finally used to assess the accuracy of the classification.

Existing vegetation indices and red-edge techniques have been widely used for the assessment of vegetation status and vegetation health from remote-sensing instruments. This study proposed and applied optimized Airborne Imaging Spectrometer for Applications (AISA) airborne hyperspectral indices in assessing and mapping stressed oil palm trees. Six vegetation indices, four red-edge techniques, a standard supervised classifier and three optimized AISA spectral indices were compared in mapping diseased oil palms using AISA airborne hyperspectral imagery. The optimized AISA spectral indices algorithms used newly defined reflectance values at wavelength locations of 734 nm (near-infrared (NIR)) and 616 nm (red). The selection of these two bands was based on laboratory statistical analysis using field spectroradiometer reflectance data. These two bands were then applied to the AISA airborne hyperspectral imagery using the three optimized algorithms for AISA data. The newly formulated AISA hyperspectral indices were D2 = R ₆₁₆/R ₇₃₄, normalized difference vegetation index a (NDVIa) = (R ₇₃₄ – R ₆₁₆)/(R ₇₃₄ + R ₆₁₆) and transformed vegetation index a (TVIa) = ((NDVIa + 0.5)/(abs (NDVIa + 0.5)) × [abs (NDVIa + 0.5)]^1/2. The classification results from the optimized AISA hyperspectral indices were compared with the other techniques and the optimized AISA spectral indices obtained the highest overall accuracy. D2 and NDVIa obtained 86% of overall accuracy followed by TVIa with 84% of overall accuracy.

Field spectroscopy is a rapid and non-destructive analytical technique that may be used for assessing plant stress and disease. The objective of this study was to develop spectral indices for detection of Ganoderma disease in oil palm seedlings. The reflectance spectra of oil palm seedlings from three levels of Ganoderma disease severity were acquired using a spectroradiometer. Denoizing and data transformation using first derivative analysis was conducted on the original reflectance spectra. Then, comparative statistical analysis was used to select significant wavelength from transformed data. Wavelength pairs of spectral indices were selected using optimum index factor. The spectral indices were produced using the wavelength ratios and a modified simple ratio method. The relationship analysis between spectral indices and total leaf chlorophyll (TLC) was conducted using regression technique. The results suggested that six spectral indices are suitable for the early detection of Ganoderma disease in oil palm seedlings. Final results after regression with TLC showed that Ratio 3 is the best spectral index for the early detection of Ganoderma infection in oil palm seedlings. For future works, this can be used for the development of robust spectral indices for Ganoderma disease detection in young and mature oil palm using airborne hyperspectral imaging.

Hyperspectral remote sensing has been applied to many agriculture applications such as crop stress and diseases, however, the potential use of this technique for disease detection in oil palms under field conditions is unknown. Hence, this research was conducted to investigate the applicability and feasibility of using a portable high resolution hyperspectral remote sensing instrument to identify and discern differences in spectral reflectance of oil palm leaves with and without Ganoderma infections. This study used the reflectance and first derivative spectra as input data in selecting the best bands that can discriminate between the healthy and the two levels of Ganoderma-infected oil palms. The results indicate that spectral derivative data have more bands that show significant differences between the classes than the use of reflectance data only. The preliminary results indicate that by using hyperpsectral data, it is possible to detect the presence of Ganoderma infection in oil palm but early detection is still very challenging.

In oil palm plantations, Basal Stem Rot (BSR) disease has been acknowledged by many researchers worldwide to be a core and consistent danger. Given this problem, the early evaluation of the disease is necessary to prevent any infection ranging from immature to mature levels within the oil palms. The main objective of this research study is to highlight several sensors and methods that are applied for disease detection at each level. This comprehensive review utilises the most recent geospatial technology together with other advanced technologies to study BSR disease remotely, based on assessments of multispectral data, hyperspectral data, electrical properties, Terrestrial Laser Scanning (TLS), spatial mapping, Intelligent Electronic Nose (e-nose), tomography techniques and RADAR sensors. This paper identifies and explains the important techniques and methods for detecting BSR disease for any researcher who has a significant interest in specifically monitoring the development of BSR disease.

High resolution field spectroradiometers are important for spectral analysis and mobile inspection of vegetation disease. The biggest challenges in using this technology for automated vegetation disease detection are in spectral signatures pre-processing, band selection and generating reflectance indices to improve the ability of hyperspectral data for early detection of disease. In this paper, new indices for oil palm Ganoderma disease detection were generated using band ratio and different band combination techniques. Unsupervised clustering method was used to cluster the values of each class resultant from each index. The wellness of band combinations was assessed by using Optimum Index Factor (OIF) while cluster validation was executed using Average Silhouette Width (ASW). 11 modified reflectance indices were generated in this study and the indices were ranked according to the values of their ASW. These modified indices were also compared to several existing and new indices. The results showed that the combination of spectral values at 610.5nm and 738nm was the best for clustering the three classes of infection levels in the determination of the best spectral index for early detection of Ganoderma disease.