E-Locus - Institutional Repository of the University of Crete - Μελέτη εφαρμογής της οπτικής φασματοσκοπίας σε συνδυασμό με μεθόδπυς πολυπαραγοντικής στατιστικής ανάλυσης στον προσδιορισμό της προέλευσης και της ποιότητας προϊόντων αγροδιατροφής

Home Μελέτη εφαρμογής της οπτικής φασματοσκοπίας σε συνδυασμό με μεθόδπυς πολυπαραγοντικής στατιστικής ανάλυσης στον προσδιορισμό της προέλευσης και της ποιότητας προϊόντων αγροδιατροφής

Results - Details

[Add to Basket]

Identifier

000428378

Title

Μελέτη εφαρμογής της οπτικής φασματοσκοπίας σε συνδυασμό με μεθόδπυς πολυπαραγοντικής στατιστικής ανάλυσης στον προσδιορισμό της προέλευσης και της ποιότητας προϊόντων αγροδιατροφής

Alternative Title

Application of spectroscopic techniques coupled with multivariate statistical analysis for the origin discrimination and quality control of agrofoods

Author

Κοντζεδάκη, Ρενάτε Γ.

Thesis advisor

Ρίζος, Απόστολος

Reviewer

Βελεγράκης, Μιχάλης
Σπύρος, Απόστολος

Abstract

Food quality control is often based on modern and sophisticated analytical methods and processes that are costly, time-consuming and require specialized personnel. At the Institute of Electronic Structure and Laser of FORTH, we employ an alternative approach in order to investigate the rapid determination of the characteristic substances contained in food samples, with little or no processing. For this study, optical spectroscopic techniques [Absorption (UV-Vis-NIR), Fluorescence and Raman spectroscopy] are applied. This way, we can record the optical spectrum that is the characteristic “fingerprint” and reflects the chemical composition of a sample. Different samples have a different fingerprint and the detected differences or similarities, coupled with multivariate statistical analysis methods (Principal Component Analysis–PCA, Partial Least Squares Discriminant Analysis PLS-DΑ and PLS-regression), are used for origin and quality control. Within the framework of the present thesis, Greek wine varieties (Asyrtiko, Vilana, Kotsifali, Liatiko and Mantilari) from three wineries were studied. Analysis of samples was implemented during their maturation in different types of containers at different stages of 0, 3, 6 and 9 months. Furthermore, spectral data were correlated with data obtained by use of other analytical methods. Efficient mathematical models were constructed that were capable of successfully estimating the content of wine in phenolics, anthocyanins and tannins. Finally, samples of Cretan olive oil were investigated, that were obtained from the Pancretan Olive Oil Competition and had been graded by a trained team of testers using human senses. The samples were discriminated based on their quality (high and low quality) and crop type (organic and conventional cultivation).

Language

Greek

Subject

Fluorescence

Olive Oil

Partial Least Squares Discriminant Analysis

Partial Least Squares regression

Principal Component Analysis

Raman

UV-VIS

Wines

Ανάλυση Διακριτότητας Μερικών Ελαχίστων Τετραγώνων