Comparison of Hyperspectral Imaging and Near-Infrared Spectroscopy to Determine Nitrogen and Carbon Concentrations in Wheat

Hyperspectral imaging (HSI) is an emerging rapid and non-destructive technology that has promising application within feed mills processing plants in poultry other intensive animal industries. HSI may be advantageous over near infrared spectroscopy (NIRS) as it scans entire samples, which enables compositional gradients sample heterogenicity to visualised analysed. This study was a preliminary investigation compare the performance of with NIRS for quality measurements ground samples Australian wheat identify most important spectral regions predicting carbon (C) nitrogen (N) concentrations. In total, 69 were scanned using (400–2500 nm), two cameras operated 400–1000 nm (VNIR) 1000–2500 (SWIR) regions. Partial least square regression (PLSR) models used correlate C N concentrations 63 calibration their reflectance, 6 additional testing models. The accuracy predictions (full spectra) similar or slightly higher than those (NIRS Rc2 = 0.90 0.96 vs. 0.97 0.97). region prediction identified reflectance 400–550 R2 0.93 RMSE 0.17% set 0.86, 0.21% ratio deviation (RPD) 2.03 test set. included 1451–1600 nm, 1901–2050 2051–2200 providing ranging from 0.91 0.93, 0.06% 0.07% sets, 0.99, RPD 3.47 3.92 sets. comparable possibly due larger statistical population (larger number pixels) provided, despite fact had smaller range compared NIRS. addition, enabled visualising variability samples. Therefore, multifunctional tool poses many potential applications data collection assurance plants. ability more accurately measure visualise properties ingredients economic benefits therefore development this warranted.