Matrix data processing method and device and storage medium

Abstract

The embodiment of the invention discloses a matrix data processing method and device and a storage medium. The method comprises the following steps: determining a first training matrix from a first spectrum matrix associated with a host, and determining a second training matrix from a second spectrum matrix associated with a slave; performing data preprocessing on the first training matrix and thesecond training matrix to obtain a first processing matrix and a second processing matrix; performing fixed half-window expansion on the second processing matrix to obtain a first expansion matrix, and performing feature reconstruction on the first expansion matrix to obtain a first feature matrix; training an initial regression model according to the first processing matrix and the first featurematrix to obtain a target regression model carrying a transfer coefficient matrix F; and obtaining a third spectral matrix associated with the slave, and predicting to obtain a target spectral matrixassociated with the host according to the third spectral matrix and the target regression model. By adopting the embodiment of the invention, the cross-equipment transfer capability of the sample canbe improved.

Description

technical field [0001] The present application relates to the technical field of near-infrared spectral analysis, and in particular to a method, device and storage medium for matrix data processing. Background technique [0002] With the continuous improvement of people's living standards, people's requirements for fruit quality are also getting higher and higher. Quality inspection and grading of post-harvest fruits have become an essential part of standardized fruit processing. Due to the advantages of fast, non-destructive, and online detection and analysis, near-infrared spectroscopy is widely used in the detection of fruit sugar content. [0003] At present, the industry usually uses the standard normal transformation algorithm to preprocess the collected near-infrared spectral data to eliminate the influence of solid particle size, surface scattering and optical path changes on the diffuse scattering spectrum of near-infrared light. However, the deviation standard cau...

AI Technical Summary

Problems solved by technology

However, the deviation standard caused by instrument factors will reduce the accuracy of the sample after migration between different devices or even fail, and the standard normal transformation algorithm cannot effectively learn the data deviation caused by the deviation of the device itself, so that the migration effect of the sample between different devices is poor

Images