[Fast discrimination of edible vegetable oil based on Raman spectroscopy].

Area of Science:

• Analytical Chemistry
• Food Science
• Spectroscopy

Background:

• Accurate identification of edible vegetable oils is crucial for quality control and consumer safety.
• Traditional methods for oil classification can be time-consuming and may require extensive sample preparation.
• Raman spectroscopy offers a rapid, non-destructive technique for chemical analysis.

Purpose of the Study:

• To develop a novel, rapid method for discriminating edible vegetable oils using Raman spectroscopy.
• To establish a classification model based on specific spectral features indicative of oil properties.
• To compare the effectiveness of the new method with Principal Component Analysis (PCA) for oil classification.

Main Methods:

• Edible vegetable oil samples were analyzed using Raman spectroscopy.
• Raw spectra underwent baseline correction and normalization.
• Two characteristic peaks related to the unsaturated degree were selected as feature vectors.
• Class centers were calculated, and Euclidean distances were used for classification of unknown samples.

Main Results:

• The novel feature extraction method demonstrated improved clustering of oil classes compared to PCA.
• A significantly larger class distance was observed with the new method, indicating better separation.
• Experimental results with 43 samples across seven classes validated the method's effectiveness.
• The classification model accurately discriminated unknown edible vegetable oil samples.

Conclusions:

• The developed Raman spectroscopy method provides a rapid and accurate approach for edible vegetable oil discrimination.
• The feature extraction technique based on characteristic unsaturated peaks enhances classification performance.
• This method holds potential for routine quality control and authentication of vegetable oils.