The aim of this thesis was to verify and evaluate the possibility of using NIR spectroscopy usage in food products quality control. The quantitative calibrations were developed to determinate of chemical parameters of sheep, goat and cow milk, honey, coffee and another food and raw materials. Food samples were evaluated by the FT-NIR Antaris spectroscope in a spectral resolution of 10 000 to 4 000 cm-1 with the possibilities of samples measurement in reflectance, transmitance or interactance mode. Calibration models were created by using a Partial Least Square algorithm (PLS) and also the cross-validation was used for the verification of the quality prediction of calibrations. Very good calibration were developed in most cases, and though some were less accurate calibrations are applicable to at least approximate determination of values. The ability of FT-NIR spectroscopy to classify sensed spectrum into predefined classes and thereby determine the differences between groups of samples was verified with a qualitative evaluation of products using with discriminant analysis. A suggestibility verification of the device was realized with cow's milk, which was measured at various temperatures. It was confirmed that the NIR spectroscopy is useful for the chemical composition and food quality evaluation.
| Identifer | oai:union.ndltd.org:nusl.cz/oai:invenio.nusl.cz:425065 |
| Date | January 2016 |
| Creators | Dvořák, Lukáš |
| Source Sets | Czech ETDs |
| Language | Czech |
| Detected Language | English |
| Type | info:eu-repo/semantics/doctoralThesis |
| Rights | info:eu-repo/semantics/restrictedAccess |
Page generated in 0.0021 seconds