Abstract
Target partial least squares (PLS) is applied to Fourier transform infrared–attenuated total reflection (FT-IR-ATR) hyperspectral images of plant leaf surface treated with crop protection products. Detection of active ingredient is demonstrated at application rates of 50 g active ingredient per hectare. This sensitivity could not be achieved without the application of multivariate analysis. Quantitative information appears to be easily recovered through analysis of combined images with known and unknown amounts of active ingredient.
Keywords
Get full access to this article
View all access options for this article.
