NIR Spectroscopy and NMF Algorithm for Identification of Oil Pollutants in Water

Oil pollutants is one of the major pollution sources in water. Accurate, rapid, and convenient detection method of oil pollutants in water has very important theoretical value and practical significance. The combination of near-infrared spectroscopy (NIR) and chemometrics is ideal for such a situation. NIR spectroscopy is a powerful and effective technique. traditional NIR methods do not take full account of the absorbance data non-negative characteristics, resulting in the analysis lack of reasonable explanation. In this paper, the qualitative discriminate method of single species oil contaminants based on nonnegative matrix factorization feature extraction combined with support vector machine classification algorithm is studied. Non-negative matrix factorization algorithm and support vector machine classifier parameters on classification accuracy are discussed in depth to optimize NIR qualitative classification model. The present method has a good identification effect and strong generalization ability , and can work as a new method for rapid identification of oil pollutants in water. Keywords-NIR Spectroscopy; Non-negative matrix factorization; oil pollutans;; Support Vector Machine; Genetic

[1]  B. Lavine,et al.  Classification of the waxy condition of durum wheat by near infrared reflectance spectroscopy using wavelets and a genetic algorithm , 2014 .

[2]  Rocco Rongo,et al.  2.7 Genetic Algorithms, Optimization, and Evolutionary Modeling , 2013 .

[3]  Jacek M. Zurada,et al.  Review and performance comparison of SVM- and ELM-based classifiers , 2014, Neurocomputing.

[4]  Yong Shi,et al.  Prediction of Customer Attrition of Commercial Banks based on SVM Model , 2014, ITQM.

[5]  Dongsup Kim,et al.  LinkNMF: identification of histone modification modules in the human genome using nonnegative matrix factorization. , 2013, Gene.

[6]  Xin-She Yang,et al.  Chapter 5 – Genetic Algorithms , 2014 .

[7]  J. Esteves,et al.  An Assessment of Oil Pollution in the Coastal Zone of Patagonia, Argentina , 2007, Environmental management.

[8]  K. Solomon,et al.  Detecting oil sands process-affected waters in the Alberta oil sands region using synchronous fluorescence spectroscopy. , 2009, Chemosphere.

[9]  Jim Jing-Yan Wang,et al.  Feature selection and multi-kernel learning for adaptive graph regularized nonnegative matrix factorization , 2015, Expert Syst. Appl..

[10]  A. Mouazen,et al.  On-line measurement of soil properties without direct spectral response in near infrared spectral range , 2013 .

[11]  Patrik O. Hoyer,et al.  Non-negative Matrix Factorization with Sparseness Constraints , 2004, J. Mach. Learn. Res..

[12]  Andri Mirzal,et al.  A convergent algorithm for orthogonal nonnegative matrix factorization , 2014, J. Comput. Appl. Math..

[13]  M. de la Guardia,et al.  Classification of persimmon fruit origin by near infrared spectrometry and least squares-support vector machines , 2014 .