A simple and quick method to detect adulterated sesame oil using 3D fluorescence spectra

Area of Science:

• Analytical Chemistry
• Food Science
• Spectroscopy

Background:

• Adulteration of sesame oil poses significant risks to consumers and market integrity.
• Rapid and accurate identification methods are crucial for ensuring food safety and quality.

Purpose of the Study:

• To develop a simple, fast, and real-time model for identifying adulterated sesame oil (ASO).
• To leverage 3D fluorescence spectra combined with wavelet moments (WMs) for effective feature extraction and classification.

Main Methods:

• Wavelet multiresolution decomposition (WMRSD) was employed to reduce noise and data volume, enhancing model stability and real-time performance.
• Wavelet moments (WMs) were utilized for feature extraction from 3D fluorescence spectra.
• Hierarchical clustering and Dunn's validity index (DVI) were used to evaluate the effectiveness of WMs for ASO identification.

Main Results:

• Wavelet moments effectively extracted key features from 3D fluorescence spectra, as confirmed by hierarchical clustering.
• The study identified optimal WMs for distinguishing adulterated sesame oil.
• The developed model demonstrated simplicity, speed, and potential for online application.

Conclusions:

• The combination of 3D fluorescence spectra and wavelet moments provides a robust method for identifying adulterated sesame oil.
• This approach offers a valuable reference for the identification and adulteration detection of vegetable oils.
• The model's real-time capability and expandability to online measurements are significant advantages for industrial application.