Discrimination of different type of meats using laser induced breakdown spectroscopy and chemometric techniques
Laser-induced breakdown spectroscopy (LIBS) is an analytical technique used for the identification of elements by analysing the emission line spectrum from samples. In this research, the possibility of classification of raw meat species based on emission spectra by using laser induced breakdown spec...
Saved in:
| 主要作者: | |
|---|---|
| 格式: | Thesis |
| 语言: | English |
| 出版: |
2015
|
| 主题: | |
| 在线阅读: | http://eprints.utm.my/id/eprint/53693/25/NurhidayuShahamiMFS2015.pdf |
| 标签: |
添加标签
没有标签, 成为第一个标记此记录!
|
| 总结: | Laser-induced breakdown spectroscopy (LIBS) is an analytical technique used for the identification of elements by analysing the emission line spectrum from samples. In this research, the possibility of classification of raw meat species based on emission spectra by using laser induced breakdown spectroscopy (LIBS) and chemometric techniques such as principal component analysis (PCA) and support vector machine (SVM) were implemented. An experimental setup was developed using Q-Switched Nd:YAG laser operating at 1064nm (208mJ per pulse) and a spectrometer connected to a fiber optic in order to collect the atomic emission. Different types of muscle tissues (beef, mutton, pork, fish, and chicken) were prepared as samples for the ablation process and the procedure for pork sample followed a specific guideline. The LIBS experiment was able to detect the elements in the meat samples such as magnesium, iron, calcium, sodium, carbon, nitrogen, and hydrogen. The raw spectra data were preprocessed and grouped into six datasets for PCA and SVM analysis. Standard ratio combination dataset showed the best result of PCA with variance of 99.8% which were later used for SVM classification. In SVM classification, the maximum accuracy of 89.33% was achieved by using a splitting ratio of 70:30 and linear kernel. The results obtained suggest a successful classification on the target tissues with high accuracy. This is valuable for an automatic discrimination in food analysis. |
|---|