Dominique Van de Sompel, Ellis Garai, Cristina Zavaleta and Sanjiv Sam Gambhir A comparison of noise models in a hybrid reference spectrum and principal components analysis algorithm for Raman spectroscopy Journal of Raman Spectroscopy 44
We compare three different noise models for use in a hybrid reference spectrum and principal components algorithm, namely an unweighted Gaussian, a Poisson, and a weighted Gaussian noise model. We demonstrate that the Poisson noise model is more accurate than the unweighted and weighted Gaussian noise models when the only signal variability is zero-mean random noise. In the presence of variations in the mean component spectra, however, modeling such variations is found to be more important to maximizing the accuracy of concentration estimates than optimizing the particular noise model used.
Complete the form below and we will send an e-mail message containing a link to the selected article on your behalf