Discriminant analysis of olive oil mill wastes using spectroradiometers in the visible and near infrared part of the spectrum

Abstract

This paper aims to introduce an index which is able to assist the recognition of Olive Oil Mill Wastes (OOMW) disposal areas using multispectral satellite images. For this purpose several ground spectroradiometric measurements in the range of 450 - 900 nm have been collected at a variety of targets. The samples were initially resampled to the GeoEye-1 sensor using the appropriate Relative Spectral Response Filter (RSR) and then scatter plots and spectral signatures were projected. Then an evaluation of nine widely used vegetation indices was performed. This analysis indicated that, Perpendicular Vegetation Index (PVI), Transformed Soil Adjusted Vegetation Index (TSAVI) and Soil and Atmospherically Resistant Vegetation Index (SARVI) are the most efficient indices to enhance the detection of Olive Mill Waste disposal areas. Moreover, correlation as well as separability analysis have shown than blue and near infrared (NIR) spectral bands are the most suitable for interpretation purposes. The Olive Mill Waste index was applied also to GeoEye-1 in different cases studies in the island of Crete.