Unlock instant, AI-driven research and patent intelligence for your innovation.

A method for expand spectral database based on generate antagonistic network

An extension method and spectral data technology, applied in biological neural network models, neural learning methods, instruments, etc., can solve the problems of time increase, human and material resources, etc.

Inactive Publication Date: 2019-03-22
BEIJING INSTITUTE OF TECHNOLOGYGY
View PDF12 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This greatly increases the time required to obtain sufficient spectra, thereby increasing manpower and material resources

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 method for expand spectral database based on generate antagonistic network
  • A method for expand spectral database based on generate antagonistic network
  • A method for expand spectral database based on generate antagonistic network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0041] Taking the LIBS spectrum of CL-20, a typical dangerous high explosive, as an example, the application of the laser-induced breakdown spectroscopy and Raman spectral data expansion modeling method based on generative adversarial networks in LIBS spectroscopy is illustrated.

[0042] Step 1. Collect LIBS spectra

[0043] build as figure 2 The laser-induced breakdown spectroscopy experimental device shown, in which 1 is a laser, 2 is a mirror, 3 is a focusing lens, 4 is a laser-induced plasma, 5 is a photodetector, 6 is a three-dimensional sample stage, 7 is an optical fiber, 8 Delay device DG535, 9 is a spectrometer, 10 is a computer.

[0044] Adhere the CL-20 powder to the surface of the glass slide with double-sided adhesive, place it on the sample stage 6, and collect the LIBS spectrum. The laser uses an Nd:YAG laser with a fundamental frequency of 1064 nm output wavelength and a frequency of 1 Hz. Adjust the three-dimensional sample stage so that each The collectio...

Embodiment 2

[0063] Taking the Raman spectrum of CL-20, a typical dangerous high explosive, as an example, the application of laser-induced breakdown spectroscopy based on generative adversarial network and Raman spectral data expansion modeling method in Raman spectroscopy is illustrated.

[0064] Step 1. Collect Raman spectrum

[0065] build as figure 2 In the Raman spectroscopy experimental device shown, 11 is a hand-held semiconductor laser and receiver, 12 is a sample bottle, and 13 is a computer of an integrated Raman spectrometer.

[0066] Put the CL-20 powder in a sample bottle, use a hand-held semiconductor laser and a receiver to align with the side wall of the sample bottle, and measure directly through the sample bottle. The laser output is a continuous laser with a wavelength of 785nm, and the integration time is 1125ms. After removing the spectral data of the first few rounds of focus adjustment, a total of 100 rounds of Raman were collected along the side wall.

[0067] S...

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

The invention relates to a spectrum database expansion method based on a generation countermeasure network, belonging to the spectrum detection technical field, which is widely applied to the expansion of near-infrared spectrum, Raman spectrum, laser-induced breakdown spectrum, fluorescence spectrum, terahertz and other spectrum databases. A small number of experimental spectra were obtained by measuring the measured samples with the spectral measurement system. The original generation spectrum with the same dimension is generated according to the dimension of the obtained experimental spectrum. A generating network G and a discriminant network D are constructed, and the generating network G and the discriminant network D are trained with each other by sharing parameter expressions. With the repetition of the above interactive training steps, the generated spectra gradually approach to the original experimental spectra with the increase of the number of interactive training. Using butnot limited to unsupervised learning clustering analysis method, the similarity between the generated spectrum and the original experimental spectrum is judged. If the requirement is not met, the interactive training step is repeated; If the requirements are met, the generated spectrum and the original spectrum are combined to form a spectral database.

Description

technical field [0001] The invention relates to a spectral database extension method based on a generative confrontation network, and belongs to the technical field of spectral detection. Background technique [0002] Spectral analysis techniques, such as Near Infrared Spectroscopy, Raman Spectroscopy, Laser-induced breakdown spectroscopy (LIBS), Fluorescence Spectroscopy and Terahertz Spectroscopy ), etc., have been widely researched and applied in the fields of explosive detection, clinical medical sample detection, cultural heritage identification, alloy processing, space exploration, and agricultural and food analysis. According to the obtained spectrum, the substance can be identified by discriminant analysis or classification. Due to the advantages of fast real-time, convenient and accurate and online in-situ detection, spectral technology will play an increasingly important role in the field of classification and identification. [0003] When using spectral data for...

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 Applications(China)
IPC IPC(8): G06K9/00G06N3/04G06N3/08
CPCG06N3/08G06V20/194G06V20/13G06N3/045
Inventor 王茜蒨腾格尔崔旭泰
Owner BEIJING INSTITUTE OF TECHNOLOGYGY