ABSTRACT
To solve the calibration transmission problem in near-infrared ( NIR) spectroscopy, a novel model transfer method, Simple Linear Regression Direct Standard-ization ( SLRDS ) , has been presented. To investigate the validity of the proposed method, a real corn sample NIR dataset was tested and the direct standardization ( DS ) method and piecewise direct standardization ( PDS ) method were involved as a comparison. Our results indicated that SLRDS can correct compressed NIR data differences among instruments and enable the user to share corn sample PLS calibration model among three instruments, at the same time it has higher prediction accuracy, fewer parameters and simpler model than DS and PDS.
ABSTRACT
In recent years, many modern spectral databases for complex materials ( such as soil, feed, forensic evidence materials, pharmaceuticals and oils) have been established based on molecular spectroscopy (UV, infrared, near infrared, Raman and fluorescence) both at home and abroad, which are playing more and more important roles in the agricultural and industrial production and science research. Spectral searching method is one of the key technologies to make full use of the molecular spectral database. This paper reviewed the progress in the basic and modified algorithm, strategy and application of molecular spectral searching methods, and put forward the scientific and technological problems that need attention and further research.
ABSTRACT
Genetic algorithms (GA), a global searching method, is applied to select wavelength variables of near infrared spectroscpy (NIR) for multivariate calibration made by partial least squares (PLS) method. Two application examples of NIR analysis show that this wavelength selection method for PLS modeling not only simplifies and optimizes calibration model but also increases the prediction ability of calibration model. The method is especially adequate for the system where only PLS is difficult to correlate.