Method for recognizing abnormal samples in near infrared spectrum analysis
A technology of near-infrared spectroscopy and abnormal samples, which is applied in the direction of material analysis, analysis materials, instruments, etc. through optical means, can solve problems such as improper methods and misoperation of chemical value data abnormalities, so as to ensure reliability, improve applicability and The effect of stability
Inactive Publication Date: 2017-04-26
NORTHEAST AGRICULTURAL UNIVERSITY
View PDF2 Cites 4 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Near-infrared spectral analysis data includes sample spectral scanning data and chemical value data of sample elements. Factors such as improper sample preparation methods, changes in surrounding environmental factors, and instrument problems will cause abnormal data in spectral data, and most of the chemical values are acquired. Obtained by chemical experiment methods, improper methods and misoperations during the experiment will also cause abnormalities in the chemical value data
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0022] The specific implementation manners of the present invention will be further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.
[0023] A method for identifying abnormal samples in near-infrared spectral analysis provided by the present invention mainly includes the following steps:
[0024] 1. Change the expression of the result of the half resampling algorithm
[0025] refer to figure 1 , in the original method, after each sampling, the abnormal samples are selected according to the confidence interval, and then the corresponding abnormal samples are accumulated and counted; instead of each sampling, the distance calculated by each sample is accumulated, and after the sampling is over, the confidence interval is selected. Samples with large distances are regarded as abnormal samples.
[0026] 2. U...
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
Login to View More Abstract
The invention discloses a method for recognizing abnormal samples in the near infrared spectrum analysis. With the combination of a half resampling algorithm and a Cook distance algorithm, the abnormal sample recognition is conducted on near infrared spectrum analysis data. Firstly, a result expression mode of the half resampling algorithm is changed to conduct the abnormal sample recognition on the spectrum data, then the abnormal sample recognition is conducted on chemical values by means of the Cook distance algorithm, each optimal confidence interval is selected through a modeling effect of partial least squares, and the combination of the two optimal confidence intervals serve as a confidence interval in the method; with regard to abnormal samples which simultaneously occur in the two methods, and if being a high leverage value, the abnormal samples are removed and otherwise reserved. According to the method, with the combination of the two mutually independent algorithms, the situation that spectroscopic anomaly and chemical value anomaly simultaneously exist or one of the spectroscopic anomaly and the chemical value anomaly exists can be handled, special abnormal samples are deeply judged, characteristic samples which are recognized due to self-characteristics are reserved, and the applicability and the stability of a model are enhanced.
Description
technical field [0001] The invention relates to the technical field of near-infrared spectrum analysis, in particular to a method for identifying abnormal samples in near-infrared spectrum analysis. Background technique [0002] Near-infrared spectroscopy has the characteristics of fast analysis speed, no damage to samples, and low cost. It has been widely used in agriculture, food, medicine and other fields. Near-infrared spectral analysis data includes sample spectral scanning data and chemical value data of sample elements. Factors such as improper sample preparation methods, changes in surrounding environmental factors, and instrument problems will cause abnormal data in spectral data, and most of the chemical values are acquired. Obtained by chemical experiment methods, improper methods and misoperations in the experimental process will also cause abnormalities in the chemical value data. The NIR spectral analysis data in different analysis processes may contain both...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G01N21/359
CPCG01N21/359
Inventor 王艳尹艳玲沈维政孙红敏李晓明
Owner NORTHEAST AGRICULTURAL UNIVERSITY