Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

A Universal Raman Spectral Feature Extraction Method for Machine Learning Substance Recognition Algorithms

A technology of Raman spectroscopy and machine learning, applied in the field of Raman spectroscopy, can solve the problems of poor feature classification and loss of peak signal strength information, and achieve the effect of improving classification accuracy and strong versatility

Active Publication Date: 2021-03-09
XIAMEN UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method often loses the intensity information of the peak signal, resulting in poor classification of the extracted features

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A Universal Raman Spectral Feature Extraction Method for Machine Learning Substance Recognition Algorithms
  • A Universal Raman Spectral Feature Extraction Method for Machine Learning Substance Recognition Algorithms
  • A Universal Raman Spectral Feature Extraction Method for Machine Learning Substance Recognition Algorithms

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0038] The following embodiments will further illustrate the present invention in conjunction with the accompanying drawings.

[0039] The present invention comprises the following steps:

[0040] The first step: automatic preprocessing of the spectrum: eliminating noise and subtracting fluorescent background;

[0041] In actual testing, Raman spectrum samples are usually expressed in the form of two-dimensional data, where the abscissa is the wave number, and the ordinate is the spectral signal intensity corresponding to the wave number. Raman spectroscopy sample collection is often affected by many factors, such as the fluorescence background (the main factor) generated by the laser, the burr peaks generated by the radiation, and the inherent noise of the instrument. In order to perform accurate substance identification on Raman spectroscopy, the influence of these factors must be eliminated as much as possible. The present invention uses an automatic spectral preprocessin...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

A general Raman spectral feature extraction method for machine learning substance identification algorithms involving Raman spectroscopy. The spectrum is automatically preprocessed; the feature vector of the spectrum is obtained. Feature extraction can be performed on any specified range of Raman spectra, and the extracted feature vectors are suitable for a variety of machine learning algorithms, which are highly versatile and not limited by target substances or test systems; noise and fluorescence background interference can be automatically removed, At the same time, information such as the position and intensity of the peak signal is retained; it can effectively identify spectra containing various target substances; it can accurately extract blank spectral features, effectively identify and accurately distinguish negative and positive samples, and better meet the actual needs of substance detection; extraction The method does not involve complex calculations and requires little storage space, so the time and space complexity is low, and it is convenient to be applied to batch processing and analysis of spectral data.

Description

technical field [0001] The invention relates to Raman spectroscopy, in particular to a general Raman spectroscopy feature extraction method for machine learning material identification algorithms. Background technique [0002] Raman spectroscopy is based on the Raman scattering effect, a vibration spectrum with molecular fingerprint information, and each substance has unique spectral information that distinguishes it from other substances. Therefore, Raman spectroscopy can detect and analyze substances, and has applications in the fields of materials, chemistry, physics, environmental protection, and life sciences. The current popular surface-enhanced Raman spectroscopy (SERS) technique [1] And the subsequent development of core-shell isolated nanoparticles enhanced Raman spectroscopy (SHINERS) technology [2] , greatly improving the sensitivity of Raman spectroscopy detection, reducing noise and background interference, and greatly improving the universality and applicabil...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00G06K9/62G01N21/65
CPCG01N21/65G06F2218/06G06F2218/08G06F2218/12G06F18/2411
Inventor 谢怡游乔贝刘国坤康怀志曾勇明孙锡龙
Owner XIAMEN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com