Method for Classifying a Noisy Raman Spectrum Based on a Wavelet Transform and a Deep Neural Network

Abstract

Because it is relatively difficult in practice to classify the Raman spectrum under baseline noise and additive white Gaussian noise environments, this paper proposes a new framework based on a wavelet transform and deep neural network for identification of noisy Raman spectra. The framework consists of two main engines. Wavelet transform is proposed as the framework front end for transforming the 1-D noise Raman spectrum to two-dimensional data. The two-dimensional data are fed to the framework back end, which is a classifier. The optimum classifier is chosen by implementing several traditional machine learning (ML) and deep learning (DL) algorithms, and we investigate their classification accuracy and robustness performances. The four chosen MLs are naive Bayes (NB), a support vector machine (SVM), a random forest (RF) and a k-nearest neighbor (KNN), and a deep convolution neural network (DCNN) was chosen as a DL classifier. Noise-free, Gaussian noise, baseline noise, and mixed-noise Raman spectra were applied to train and validate the ML and DCNN models. The optimum back-end classifier was obtained by testing the ML and DCNN models with several noisy Raman spectra (10-30 dB noise power). Based on the simulation, the accuracy of the DCNN classifier is 9% higher than that of the NB classifier, 3.5% higher than the RF classifier, 1% higher than the KNN classifier, and 0.5% higher than the SVM classifier. In terms of robustness to mixed noise scenarios, the framework with the DCNN back end showed superior performance compared with the other ML back ends. The DCNN back end achieved 90% accuracy at 3 dB SNR, while the NB, SVM, RF, and K-NN back ends required 27 dB, 22 dB, 27 dB, and 23 dB SNR, respectively. In addition, in the low-noise test dataset, the F-measure score of the DCNN back end exceeded 99.1%, and the F-measure scores of the other ML engines were below 98.7%.

Deep Learning for Raman Spectroscopy: A Review

“…The feature extraction part contains an input layer, a convolutional layer, and a ReLU layer; the classification part contains a multilayer perceptron (MLP) and a softmax output layer. Besides, Pan and his colleagues even increased the Raman data dimension from 1D to 2D by wavelet transform before classification [49,50]. In addition, a single-layer multiple-kernel-based convolutional neural network (SLMK-CNN) containing one convolutional layer with five different kernels, one flatten layer, and two fully-connected layers was created for Raman spectra obtained from porcine skin samples [51].…”

“…Examples of typical deep learning applications for Raman spectroscopy. Dong et al[37], Lee et al[38], Kirchberger-Tolstik et al[39], Maruthamuthu et al[40], Cheng et al[41], Fan et al[42], Fu et al[43], Houston et al[44], Ho et al[45], Ding et al[46], Chen et al[47], Saifuzzaman et al[48], Pan et al[49,50], Sohn et al[51], Yu et al[52], Thrift and Ragan[53], and Zhang et al[54] …”

An end‐to‐end deep learning approach for Raman spectroscopy classification

“…This information includes the wavenumber, changes of wavenumber, polarization, width, and intensity of Raman peak, which represent the composition, stress state, crystal symmetry and orientation, crystal mass, and amount of material, respectively. However, the Raman signal is easily affected by the fluorescence process, material density, environmental noise, and external light source; the resulting spectrum always shows baseline drift and is interfered by the noise signals 6 . Moreover, these noise signals can be several orders of magnitude higher than Raman scattering, seriously affecting the spectrum analysis.…”

“…However, the Raman signal is easily affected by the fluorescence process, material density, environmental noise, and external light source; the resulting spectrum always shows baseline drift and is interfered by the noise signals. 6 Moreover, these noise signals can be several orders of magnitude higher than Raman scattering, seriously affecting the spectrum analysis. Due to these limitations, preprocessing methods are required in the traditional analysis process of Raman spectroscopy to extract the critical spectral information contained in the vibrational fingerprints.…”

“…In addition, other researchers used deep convolutional neural networks for noise removal and accurate correction. 6,9,10 Based on the preprocessing of Raman spectral signal, the manual analysis process extracts the spectral signal features through feature peak extraction and variable selection. Alternatively, combine it with chemometric analysis methods or machine learning (ML) methods or deep learning (DL) methods for qualitative or quantitative analysis.…”