Click on “Download PDF” for the PDF version or on the title for the HTML version.
If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.
Integrating NIR Reflectance, Vis/NIR Interactance, and Hyperspectral Imaging Systems Data to Predict Physiological Status of Potato Tubers
Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org
Citation: Paper number 131671283, 2013 Kansas City, Missouri, July 21 - July 24, 2013. (doi: http://dx.doi.org/10.13031/aim.20131671283) @2013
Authors: Ahmed M Rady, Daniel E Guyer
Keywords: Reflectance interactance potato hyperspectral partial least squares neural networks glucose sucrose.
Abstract. Data from different spectroscopic systems (VIS/NIR interactance, and NIR reflectance) in addition to VIS/NIR hyperspectral reflectance imaging system were combined to study the effect on prediction of glucose and sucrose for two common potato cultivars (FL1879 and Russet Norkotah) in two seasons (2009 and 2011 for sliced samples and whole tubers. Results of sliced samples indicated that glucose prediction performance for FL1879 was improved from R(RPD) values of 0.93(2.56) into 0.95(3.45). Also, sucrose prediction was improved from R(RPD) values of 0.77 (1.51) into 0.95(3.78) using the three systems combined data. For Russet Norkotah, glucose prediction was improved from 0.91(2.36) into 0.98(9.37), and the sucrose prediction was improved from R(RPD) values of the combining interactance and reflectance data for sliced samples showed improvement of glucose prediction for Russet Norkotah from R(RPD) values of 0.91(2.36) to 0.98(9.3, using the interactance reflectance combined data. However, no improvement was noticed for sucrose prediction. For Whole tubers and using interactance reflectance combined data, FL1879 glucose prediction was improved from R(RPD) values from 0.82(1.74) into 0.95(3.30). Sucrose prediction was improved from 0.87(1.40) to 0.99(6.46). For Russet Norkotah, glucose prediction was improved from R(RPD) values of 0.95(3.15) into 0.98(4.99). And sucrose prediction was improved from 0.45(1.13) into 0.98(2.93). Previous results showed advantages from combining data.
(Download PDF) (Export to EndNotes)