Proceedings of the First International Conference on Information Sciences, Machinery, Materials and Energy

Near Infrared Spectroscopy Analysis Based on Support Vector Machine

Authors
Min Li, Linju Lu, Jin Cao
Corresponding Author
Min Li
Available Online July 2015.
DOI
https://doi.org/10.2991/icismme-15.2015.68How to use a DOI?
Keywords
Near Infrared Spectroscopy; Multiplicative Scatter Correction; Linear Discriminant Analysis; Support Vector Machine.
Abstract
This paper put forward a kind of qualitative identification method of tea authenticity based on near infrared spectroscopy(NIR). Authentic Zhuyeqing tea and fake Zhuyeqing tea were the research objects. Multiplicative Scatter Correction (MSC) was used to NIR data of 2 kinds of Zhuyeqing tea as a pre-processing. Principal Component Analysis (PCA) was then used to spectral data for dimensionality reduction and redundant removal. Next Linear Discriminant Analysis (LDA) was used for further feature extraction. Finally Support Vector Machine (SVM) was run for identification. Rradial basis was chosen as the support vector kernel function. On the condition of , modeling recognition effect is the best of 97%. Experiments show that the algorithm can effectively identify 2 kinds of Zhuyeqing tea.
Open Access
This is an open access article distributed under the CC BY-NC license.

Download article (PDF)

Proceedings
First International Conference on Information Sciences, Machinery, Materials and Energy
Part of series
Advances in Intelligent Systems Research
Publication Date
July 2015
ISBN
978-94-62520-67-7
ISSN
1951-6851
DOI
https://doi.org/10.2991/icismme-15.2015.68How to use a DOI?
Open Access
This is an open access article distributed under the CC BY-NC license.

Cite this article

TY  - CONF
AU  - Min Li
AU  - Linju Lu
AU  - Jin Cao
PY  - 2015/07
DA  - 2015/07
TI  - Near Infrared Spectroscopy Analysis Based on Support Vector Machine
BT  - First International Conference on Information Sciences, Machinery, Materials and Energy
PB  - Atlantis Press
SP  - 345
EP  - 348
SN  - 1951-6851
UR  - https://doi.org/10.2991/icismme-15.2015.68
DO  - https://doi.org/10.2991/icismme-15.2015.68
ID  - Li2015/07
ER  -