Spectral Normalization
Mathematical correction adjusts for the physical effects of light scattering in near-infrared spectroscopy of solid materials. Multiplicative scatter correction removes the variations in path length and light distribution caused by the texture or particle size of a paper sample. This ensures that the resulting spectra reflect the chemical composition of the material rather than its physical surface properties.
Calibration Stability
Applying this transformation simplifies the development of mathematical models that predict properties like moisture or starch content. When multiplicative scatter correction is used, the model becomes more robust against changes in the surface roughness of the board. It works by regressing each individual spectrum against a reference spectrum, typically the mean of the entire set.
The slope and intercept of this regression are then used to correct the original data points.
Measurement Precision
This technique is useful in an industrial environment where the distance between the sensor and the moving web might fluctuate. Multiplicative scatter correction allows for more accurate real-time monitoring of the production process without frequent recalibration. It helps in identifying subtle chemical changes that would otherwise be hidden by the noise of the scattered light.
The corrected data provides a cleaner input for subsequent analysis using regression techniques. Spectral stability is improved through this numerical adjustment.
