Nitrogen-rich adulterants in protein powders present sensitivity challenges to conventional combustion methods of protein determination which can be overcome by near Infrared spectroscopy (NIRS). NIRS is a rapid analytical method with high sensitivity and non-invasive advantages. This study developed robust models using benchtop and handheld spectrometers to predict low concentrations of urea, glycine, taurine, and melamine in whey protein powder (WPP). Effectiveness of scanning samples through optical glass and polyethylene bags was also tested for the handheld NIRS. WPP was adulterated up to six concentration levels from 0.5% to 3% w/w. The two spectrometers were used to obtain three datasets of 819 diffuse reflectance spectra each that were pretreated before linear discriminant analysis (LDA) and regression (PLSR). Pretreatment was effective and revealed important absorption bands that could be correlated with the chemical properties of the mixtures. Benchtop NIR spectrometer showed the best results in LDA and PLSR but handheld NIR spectrometers showed comparatively good results. There were high prediction accuracies and low errors attesting to the robustness of the developed PLSR models using independent test set validation. Both the plastic bag and optical glass gave good results with accuracies depending on the adulterant of interest and can be used for field applications.

Near Infrared Hyperspectral Imaging (NIRHSI) is an emerging technology platform that integrates conventional imaging and spectroscopy to attain both spatial and spectral information from an object. Two important problems in NIRHSI are those of data load and unserviceable pixels in the NIR sensor. Hyperspectral imaging experiments generate large amounts of data (typically > 50 MB per image), which tend to overwhelm the memory capacity of conventional computer systems. This inhibits the utilisation of NIRHSI for routine online industrial application. In general, approximately 1% of pixels in NIR detectors are unserviceable or ‘dead’, containing no useful information. While this percentage of pixels is insignificant for single wavelength imaging, the problem is amplified in NIRHSI, where > 100 wavelength images are typically acquired. This paper describes an approach for reducing the data load of hyperspectral experiments by using sample-specific vector-to-scalar operators for real time feature extraction and a systematic procedure for compensating for ‘dead’ pixels in the NIR sensor. The feasibility of this approach was tested for prediction of moisture content in carrot tissue.

No abstract

No abstract

No abstract