Sugarcane qualitative attributes determination and mapping by spectral sensors

Abstract

For some agricultural crops the characterization of product quality becomes as important as the measurement of productivity. Sugarcane is an example. The crop profitability is a result association of productivity and quality. Despite the advances in productivity measurement, including the new sensors development in harvesters, quality measurement in real time is still a distant reality. Thus, aim is to evaluate the predicting potential of sugarcane qualitative attributes by different spectroscopy techniques and to analyze this techniques potential for mapping the spatial variability of these attributes in the field. Visible and Near Infrared (visNIR) and X-Ray Fluorescence (XRF) spectroscopy techniques will be used in this work. The sensors readings based on these techniques will be carried out in an industrial quality laboratory parallel to the traditional analyzes for the same samples. In addition, readings will be made for different forms of same sample (stem, fiber and juice). The most suitable sensing techniques (or the combination of these) and sample form for qualitative attributes quantification (Brix, Pol, fibers, purity, sugar reducers and ATR) of sugarcane will be evaluated. Finally, the most appropriate technique will be used to read geo-referenced samples from a commercial field in order to analyze the potential for field application and identification of existing spatial variability for each attribute. (AU)