Detecting Adulteration in Camellia Oil Using Raman Spectroscopy
Author Information
Author(s): Liu Henan, Ma Sijia, Liang Ni, Wang Xin
Primary Institution: School of Physical Science and Technology, Tiangong University, Tianjin, China
Hypothesis
Can Raman spectroscopy combined with machine learning algorithms accurately detect adulteration in camellia oil products?
Conclusion
The study found that the best prediction accuracy for camellia oil adulteration was achieved using a PLSR model coupled with CARS-ICA feature extraction.
Supporting Evidence
- The study demonstrated that BPNN and PLSR models achieved satisfactory prediction performances for camellia–rice blended oils.
- Raman spectroscopy was found to be effective for detecting adulteration in oils with similar compositions.
- Feature extraction methods significantly influenced the prediction accuracy of the models used.
Takeaway
This study shows how scientists can use special light techniques to check if camellia oil is mixed with cheaper oils, helping keep our food safe.
Methodology
The study used Raman spectroscopy to analyze camellia oil adulteration with rice bran and corn oils, employing various machine learning and chemometric algorithms for prediction.
Limitations
The study primarily focused on binary and ternary adulteration scenarios and may not cover all possible adulterants.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website