Clustering analysis-based spectral model transferring method

Abstract

The invention discloses a clustering analysis-based spectral model transferring method. The method comprises the following steps of: (1) sample classification: a density clustering algorithm is adopted to perform clustering analysis on samples according to density distances between the spectra of the samples so as to classify the samples; (2) spectral modeling: spectral models are respectively established for samples of the same category according to a clustering analysis result; and (3) automatic model transfer: the spectrum of a to-be-detected sample is collected, and clustering analysis is carried out on the to-be-detected sample according to the method in the step (1), the category of the sample to be detected is determined, so that a spectral model to which the to-be-detected sample belongs can be determined, sample rapid measurement is carried out according to the spectral model to which the sample belongs, so that model transfer can be automatically completed. With the method disclosed by the invention adopted, spectral model transfer can be rapidly and efficiently completed. The clustering analysis-based spectral model transferring method can be used for the automatic calibration of an instrument, really realizes the rapid measurement of samples, and ensures measurement accuracy.

Description

technical field [0001] The invention relates to a spectral model transfer method, in particular to a spectral model transfer method based on cluster analysis. Background technique [0002] Spectral analysis technology is widely used in geology, metallurgy, petroleum, chemical industry, agriculture, Medicine, biochemistry, environmental protection and many other aspects. However, these applications are limited to the interior of homogeneous samples, which greatly limits the depth to which this technique can be used. [0003] To apply the spectral model of one type of sample to another type of sample, model transfer must be performed. Previous model transfer methods fall into two categories: labeled methods and unlabeled methods. Among them, the standard method uses a small-scale standard sample set to recalibrate (model), the procedure is cumbersome and time-consuming, and the effect is better; the standard-free method completely relies on the chemometric method, the proce...

AI Technical Summary

Problems solved by technology

This means that before using the spectral model of the master sample, it is necessary to understand the slave sample, that is, to determine the chemical and spectral data of a representative part of the slave sample, and the former is a tedious and time-consuming process.

Therefore, the existing model transfer has not achieved a real quick test

Images