A machine learning classification methodology for Raman Hyperspectral imagery based on auto‐encoders

Abstract

Raman spectroscopy is a very sensitive, non‐destructive optical technique that is able to characterize the composition of a material. Although it is a very useful technique, its signals are rather complex and therefore require a lot of knowledge and skills to understand their meaning. This complexity increases further, when it is combined with spatial mapping, creating a three‐dimensional dataset named Raman hyperspectral imagery. The common methodology of analysing such a dataset is to use a generic supervised classification algorithm (i.e., non‐negative least square) combined with reference data. If reference data are not available, unsupervised algorithms are used to perform spectral clusterization of similar areas. This is typically achieved using standard PCA and k‐means methods. The most common problem of processing that significantly affects the results of analysis is that a raw Raman Signal consists of three sources: Raman scattering, photoluminescence and noise. The contribution weights of those components are not standard and are changing based on the sample's chemistry, which makes it very difficult to be solved. This creates the necessity for pre‐processing methods that can split this raw signal to its components. The goals of this project are to improve the unmixing performance using AI and to achieve a successful split of the raw signal into its three components. Through this study, various unsupervised methods are explored to identify the method that fits best to our microscopic RHSI data sets. Our focus is on using autoencoders for the data analysis and on a Gaussian smoothing filter combined with polynomial fittings to split the data. Results show that the selection of a classification methodology depends greatly on the application, but when a large RHSI dataset is available, the newly proposed splitting technique combined with an autoencoder classification presents the best solution.

Raman and photoluminescence signal separation in Raman hyperspectral imagery including noise reduction

“…For a complete and highly detailed reference guide to RHSI preprocessing and the mathematical theory behind the algorithms used in this paper, we refer to the work of Goedhart. 17…”

“…Methods for general noise reduction are often based on Savitzky-Golay, 7 principle component analysis (PCA), 8 low-pass filters (LPFs), 9 or Gaussian filters. 10,11 Savitzky-Golay algorithms reduce noise based on local gradients, resulting in the incorrect removal or reduction of Raman spikes. Furthermore, these algorithms lack hyperparameters, making them inflexible.…”

“…At best, weak supervision can be employed. 17 However, these methods are highly dependent on the algorithm used to create the target data, resulting in solutions that are inherently biased toward those algorithms.…”

3D Raman mapping combined with FIB‐SEM on multiphase fluid inclusions: A tool to unravel complex phase assemblages

“…Thus, for mapping out phases often smaller than a few μm in 3D requires a different approach. Another common approach is to use a supervised classification algorithm spectrum‐by‐spectrum (i.e., pixel‐by‐pixel), based on a reference dataset, which is used through a least‐squares unmixing algorithm 32 . To achieve this, we used the CLS (classical least‐squares) method built in LabSpec6.…”

“…Another common approach is to use a supervised classification algorithm spectrum-by-spectrum (i.e., pixel-by-pixel), based on a reference dataset, which is used through a least-squares unmixing algorithm. 32 To achieve this, we used the CLS (classical least-squares) method built in LabSpec6. This method decomposes each spectrum into the chosen components, outputs the list of weights (scores) of the used reference spectra of each phase (components), and colors the pixel of each point of the spectral map accordingly (Figure S2).…”