The application of spectroscopic methods to predict sugarcane quality based on stalk cross-sectional scanning

Nawi, Nazmi Mat and Jensen, Troy and Chen, Guangnan (2012) The application of spectroscopic methods to predict sugarcane quality based on stalk cross-sectional scanning. American Society of Sugar Cane Technologists Journal, 32. pp. 16-27. ISSN 1075-6302

[img] PDF (Published Version)
Nawi_Jensen_Chen_PV.pdf

Download (183Kb)

Abstract

With the increasing adoption of Precision Agriculture (PA) technique in the sugarcane industry, there is a growing need for a reliable method of in-field quality measurement. However, current PA monitoring systems can only monitor cane yield and have no ability to measure the product quality. Thus, the purpose of this study was to evaluate the ability of the spectroscopic techniques as a rapid and non-destructive tool to predict quality properties of sugarcane in the field. Both handheld Vis/NIR (350-1075 nm) and full range (350 - 2500 nm) spectroradiometers were used to determine the quality attributes of sugarcane by scanning the cross-sectional surface of the stalk. NIR calibration models were constructed using a set of 100 stalks, each which were further cut into three sections of top, middle and bottom sections. After preprocessing treatments, Partial Least Squares (PLS) method was used to interpret spectra and to develop calibration model for sugarcane quality. The overall coefficient of determination (r2) for Brix, Pol, CCS and fibre as predicted by the Vis/NIR for all sample sections were 0.68, 0.71, 0.70 and 0.56 respectively. The corresponding r2 for Brix, Pol, CCS and fibre as predicted by the FRs were 0.76, 0.76, 0.81 and 0.68 respectively. It was found that by using Vis/NIR, the top section can achieve r2 of 0.89 for CCS prediction. The results suggested that spectroscopy based on stalk cross-sectional scanning is a feasible method for mapping and predicting sugarcane quality in the field.


Statistics for USQ ePrint 21015
Statistics for this ePrint Item
Item Type: Article (Commonwealth Reporting Category C)
Refereed: Yes
Item Status: Live Archive
Additional Information: Open access journal.
Depositing User: Mr Nazmi Mat Nawi
Faculty / Department / School: Historic - Faculty of Engineering and Surveying - Department of Agricultural, Civil and Environmental Engineering
Date Deposited: 14 Nov 2012 05:18
Last Modified: 26 Mar 2013 04:29
Uncontrolled Keywords: sugarcane; commercial cane sugar (CCS); quality attributes; NIR; spectroradiometer
Fields of Research (FOR2008): 07 Agricultural and Veterinary Sciences > 0701 Agriculture, Land and Farm Management > 070104 Agricultural Spatial Analysis and Modelling
08 Information and Computing Sciences > 0801 Artificial Intelligence and Image Processing > 080104 Computer Vision
09 Engineering > 0999 Other Engineering > 099901 Agricultural Engineering
Socio-Economic Objective (SEO2008): B Economic Development > 82 Plant Production and Plant Primary Products > 8203 Industrial Crops > 820304 Sugar
URI: http://eprints.usq.edu.au/id/eprint/21015

Actions (login required)

View Item Archive Repository Staff Only