Performance characterization of miniaturized near-infrared (nir) spectrometers for the classification of sweet and bitter almonds

Identifier:  TFM:1218

Handle: https://hdl.handle.net/20.500.11797/TFM1218

Authors:  Ahmed, Hawbeer Jamal

Abstract:
The miniaturized NIR instruments have gained more interest in recent years because of the fast and robust measurements they provide with almost no sample pretreatments. In this work, different sources of variability have been studied that may affect the performance of two portable low-cost NIR devices (SCiO and NeoSpectra). Measurement error covariance and correlation matrices were calculated and then visually inspected to evaluate the sources and structures of the errors associated with both devices and to find the optimal preprocessing technique that may result in the improvement of the models built with both devices.