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

Near infrared spectrum data single classification feature extraction method based on spatial decomposition

A technology of near-infrared spectroscopy and classification features, which is applied in the field of single-classification feature extraction of near-infrared spectroscopy data, can solve the problem of high universality, and achieve the effect of high universality, balanced feature compression and detection sensitivity reduction

Active Publication Date: 2020-12-01
WENZHOU UNIVERSITY
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The technical problem to be solved by the present invention is to provide a single-category feature extraction method based on spatial decomposition of near-infrared spectral data, which can effectively balance the problem of single-category feature compression and detection sensitivity reduction, and has high universality

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
  • Near infrared spectrum data single classification feature extraction method based on spatial decomposition
  • Near infrared spectrum data single classification feature extraction method based on spatial decomposition
  • Near infrared spectrum data single classification feature extraction method based on spatial decomposition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0026] Embodiment: a kind of near-infrared spectral data single classification feature extraction method based on spatial decomposition, comprises the following steps:

[0027] (1) According to actual needs, collect corresponding samples to construct training set and test set. The number of samples included in the training set is recorded as n1, and n1 is an integer greater than or equal to 10. The number of samples included in the test set is recorded as n2, and n2 is An integer greater than or equal to 1, using a near-infrared spectroscopy instrument to obtain the near-infrared spectroscopy data of each sample in the training set and test set;

[0028] The acquired near-infrared spectrum data of each sample is stored in the form of a row vector, and the number of columns of each sample row vector is its characteristic dimension, which is recorded as p, that is, the near-infrared spectrum data of each sample is 1 row Row vector of p columns;

[0029] The near-infrared spectr...

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 discloses a near infrared spectrum data single classification feature extraction method based on spatial decomposition. A principal part feature extraction method is introduced into single-classification feature extraction by adopting a spatial decomposition method; the method comprises the steps of extracting feature information of a target class in a main space through a principalpart analysis method, extracting feature information of an abnormal class in a supplementary space through a first-order norm, and reserving main information of the main space and the supplementary space while efficiently compressing the feature information of a global space; the method has the advantages that the problems of single classification feature compression and detection sensitivity reduction can be effectively balanced, and the universality is high.

Description

technical field [0001] The invention relates to a single-category feature extraction method for near-infrared spectrum data, in particular to a single-category feature extraction method for near-infrared spectrum data based on spatial decomposition. Background technique [0002] Near-infrared spectroscopy detection technology is a non-destructive and rapid detection technology, which has been widely used in agriculture, chemical industry, medicine, food and other fields, and has gradually been accepted by the public and officially recognized. The single-category model is established by using the near-infrared spectral data, and the qualitative analysis of the composition of the substance can be carried out. However, the near-infrared spectral data has high-dimensional properties, which seriously restricts the generalization effect of the single classification model established. Feature extraction for near-infrared spectral data to reduce the influence of high-dimensional da...

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): G06K9/00G06K9/20G06K9/62G06F17/16
CPCG06F17/16G06V30/412G06V10/143G06F18/2113G06F18/214
Inventor 陈孝敬黄光造石文蒋成玺袁雷明陈熙
Owner WENZHOU UNIVERSITY