MULTIVARIATE CALIBRATION OF NEAR INFRARED SPECTROSCOPY IN THE PRESENCE OF LIGHT SCATTERING EFFECT: A COMPARATIVE STUDY
Wang, K, Chi, G, Lau, R and Chen, T (2011) MULTIVARIATE CALIBRATION OF NEAR INFRARED SPECTROSCOPY IN THE PRESENCE OF LIGHT SCATTERING EFFECT: A COMPARATIVE STUDY ANALYTICAL LETTERS, 44 (5). 824 - 836. ISSN 0003-2719
Available under License : See the attached licence file.
Official URL: http://dx.doi.org/10.1080/00032711003789967
When analyzing heterogeneous samples using spectroscopy, the light scattering effect introduces non-linearity into the measurements and deteriorates the prediction accuracy of conventional linear models. This paper compares the prediction performance of two categories of chemometric methods: pre-processing techniques to remove the non-linearity and non-linear calibration techniques to directly model the non-linearity. A rigorous statistical procedure is adopted to ensure reliable comparison. The results suggest that optical path length estimation and correction (OPLEC) and Gaussian process (GP) regression are the most promising among the investigated methods. Furthermore, the combination of pre-processing and non-linear models is explored with limited success being achieved.
|Additional Information:||This is an electronic version of an article published in ANALYTICAL LETTERS, 44(5), 824-836 (2011). ANALYTICAL LETTERS is available online at: http://www.tandfonline.com/loi/lanl20.|
|Uncontrolled Keywords:||Science & Technology, Physical Sciences, Chemistry, Analytical, Chemistry, Light scattering correction, Non-linear multivariate calibration, Pre-processing, ARTIFICIAL NEURAL-NETWORKS, SUPPORT VECTOR MACHINES, LEAST-SQUARES, REFLECTANCE SPECTRA, BAYESIAN FRAMEWORK, SIGNAL CORRECTION, REGRESSION, TRANSMITTANCE, QUANTITATION, ABSORPTION|
|Divisions:||Faculty of Engineering and Physical Sciences > Chemical and Process Engineering|
|Deposited By:||Symplectic Elements|
|Deposited On:||22 Aug 2012 12:02|
|Last Modified:||25 Apr 2013 13:48|
Repository Staff Only: item control page