Raman spectrum band region segmentation method

Abstract

The invention relates to a Raman spectrum band region segmentation method, which comprises the following steps of: preprocessing Raman spectrum data, including SG smoothing and baseline removal; marking important Raman spectrum band regions by experts according to Raman characteristic peaks, and constructing a training set and a test set of a Raman spectrum segmentation model; and establishing a Raman spectrum important spectrum band segmentation model by using a deep learning algorithm, using a 1D convolutional neural network with residual edges to extract Raman spectrum characteristic information in the model, and using an anchor point structure to predict an important spectrum band region. The method can efficiently and accurately segment the important spectrum band region of the Raman spectrum data, is high in practicability, and can accurately position the important characterization region of the Raman spectrum.

Description

technical field [0001] The present invention relates to the field of spectral informatics, in particular to the field of optical detection technology, and specifically refers to a method for segmenting Raman spectrum bands based on a deep learning model. Background technique [0002] Raman spectroscopy analysis technology has been widely used in various fields, such as rock composition detection, biomedical detection, sewage composition detection, etc. In the Raman spectroscopy analysis technology, the Raman spectroscopy data analysis method is one of the important links. [0003] At present, Raman spectral data analysis methods mainly focus on identifying the type of the measured substance, and its identification methods mainly include principal component analysis (PCA), linear discriminant analysis (LDA), and support vector machine (SVM) based on machine learning. However, the current detection method cannot segment the important band region of Raman spectrum. Therefore, ...

AI Technical Summary

Problems solved by technology

However, the current detection method cannot segment the important band region of the Raman spectrum. Therefore, the identified Raman spectrum result cannot be accurately positioned to the band region corresponding to the characteristic peak, and thus the detection result of the model cannot be represented.

Images