Multivariable correction characteristic wavelength selection method based on minimum correlation coefficient

Abstract

The invention discloses a multivariable correction characteristic wavelength selection method based on a minimum correlation coefficient, and aims to solve the problem of an existing wavelength selection method. The method comprises the following steps: performing S-G first-order derivative processing on the spectral data set X; calculating absolute values of correlation coefficients among the column vectors; obtaining correlation coefficient matrix R, calculating an average value and a standard deviation of other elements except diagonal lines in each column in the correlation coefficient matrix R; selecting a correlation coefficient average value and a standard deviation threshold value pair; forming a to-be-selected wavelength set S; sorting the wavelengths of the S set to obtain a setS'; gradually adding a wavelength variable to establish an MLR model; calculating the RMSEV value of each model, taking the variable subset corresponding to the minimum RMSEV value as the characteristic wavelength under S, selecting the next threshold pair, repeating the above steps, and finding the corresponding minimum RMSEV value and the corresponding characteristic wavelength under all characteristic wavelength sets. According to the variable selection method, redundancy is reduced to the maximum extent, the principle is simple, and implementation is easy.

Description

technical field [0001] The invention relates to the field of near-infrared spectrum wavelength selection, in particular to a multivariate correction characteristic wavelength selection method based on the minimum correlation coefficient. Background technique [0002] In recent years, near-infrared spectroscopy has been widely used in petrochemical, pharmaceutical, environmental, clinical, agricultural, food and biomedical fields. The near-infrared spectral region (800-2500nm) is mainly composed of double frequency and combined frequency absorption peaks of hydrogen-containing groups. The absorption intensity is weak and the sensitivity is low. There will be deficiencies such as multicollinearity or too many non-informative variables. Selecting the characteristic wavelengths for the full spectrum is to reduce data redundancy and multicollinearity, which can improve the prediction accuracy of the model and simplify the complexity of the model. [0003] Common variable select...

AI Technical Summary

Problems solved by technology

Among them, the successive projections algorithm (successive projections algorithm, SPA) is a wavelength selection algorithm that minimizes collinearity among variables through vector projection analysis, but its principle is not easy to understand, and its operation is more complicated. The rest of the algorithms also have variables. The problem of collinearity among them, people are also doing related research

Images