New algorithm developed to improve the transfer e

image: Original and average spectra of healthy and normal wheat and maize grains.
to see After

1 credit

Recently, a research team from the Hefei Institutes of Physical Sciences (HFIPS) of the Chinese Academy of Sciences (CAS) developed a new algorithm in the direction of near-infrared spectroscopy technology, which is used to improve the efficiency of transfer of near infrared qualitative analysis. models between instruments.

The corresponding results have been published online in Infrared physics and technology.

Near infrared spectroscopy (NIRS) is a fast and non-destructive detection technology. Calibration models are the key to NIRS analysis, and the accuracy of model transfer between instruments determines the effectiveness of the popularization and application of this technology. To ensure that the predictive performance of models is not affected when transferred between instruments, new algorithms and calibration techniques must be continuously developed. In previous studies, researchers have mainly focused on quantitative model transfer in the near infrared, while there have been few reports of qualitative model transfer.

To address this issue, the team comparatively studied various transfer algorithms with near-infrared identification of defective grains in wheat and maize grains as examples, aiming to optimize the performance of qualitative models in the near. infrared during the transfer of different instruments, and to improve the robustness of the NIRS prediction.

The research team proposed a correlation analysis-based wavelength selection (CAWS) method in a previous study to improve the transfer efficiency of near-infrared quantitative models by screening bands of stable and coherent waves between instruments.

This time, the researchers further improved the CAWS algorithm to make it also applicable to qualitative discrimination models.

The results show that the validation Matthews correlation coefficients of the CAWS-optimized wheat and maize discriminant models are 0.718 and 1, respectively, ranking second and first under various processing conditions of the algorithm, which verifies the effectiveness of the proposed method.

This study proposes an algorithm to improve the transfer efficiency of near-infrared qualitative models between instruments, which is beneficial for the popularization and application of near-infrared spectroscopy.


Warning: AAAS and EurekAlert! are not responsible for the accuracy of press releases posted on EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.

Comments are closed.