The Discrimination of Fluorescence Spectra of Phytoplankton for Environment Protection Based on the PCA and SVM

Abstract:

Issues of environmental protection and sustainable development are gaining an increasing importance in everyday life, and nowhere is this more than in the field of Materials Science and Engineering. The alga is the most common phytoplankton, identifying them can estimate the community structure and distribute status of ecosystem in the sea area and realize the inspecting and comprehensive father of sea. In this paper, the three dimension fluorescence spectra and principal component analysis method is combined to identify the ocean phytoplankton. Aiming at the east China sea, adopt the selection of common seaweed three-dimensional fluorescence spectrum of first principal component scores spectrum as bacillariophyta and pyrrophyta identification features diatoms and spectrum, established the phytoplankton fluorescence features spectrum library. On this basis, the SVM classifier is used to identify the kinds of the phytoplankton. The accuracy of the experimental results recognition is for 95%.

• Periodical: Advanced Materials Research (Volume 600)
• Pages: 63-66
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.600.63
• Online since: November 2012
• Authors: Hai Yan Chen, Ni Yang, Feng Xia Han
• Keywords: Environment Protection, Fluorescence Spectra, Phytoplankton, Principal Component Analysis (PCA), Support Vector Machine (SVM)

References

[1] Su Rongguo, Liang Shengkang, Hu Xupeng, Wang Xiulin, Discrimination of 6 toxic red tide algae occurred inEast China Sea by 3D fluorescence spectra, MARINE ENVIRONMENTAL SCIENCE, vol 27, 2008, pp.265-268.

[2] Su Rongguo, Liang Shengkang, Zhu Chenjian, Wang Xiulin, Fluor escence Discr iminat ion Technology of Bacillariophyta and Pyrrophyta, Environmental Science & Technology, vol 31, 2008, pp.52-55.

[3] Sun Chongxuan, Human Face Recognition with PCA Method, Journa l of Yunnan Un ive rsity of Nationalities( Natura l Sc iences Ed ition), vol 19, 2010, pp.439-443.

[4] Tian Yanan, Lei Hongw ei, Wang Xu, An Automatic Navigation System Based on SVM Image Segmentation for the Blind, Journal of Northeastern University(Natural Science), vol 31, 2010, pp.1681-1684.

[5] Zhang Min, Li Taoshen, Zhong Shuying, Implementation of principal components analysis method based on Matlab, Journal of Guangxi University, vol 30(s), 2005, pp.74-77.

[6] R.P.W. Duin, P. Juszczak, D. de Ridder, PR-Tools, a Matlab toolbox for pattern recognition, http: /www. prtools. org, (2004).