This chapter is about the application of latent variable-based regression methods on hyperspectral images. It is an applied chapter, and no new PLS algorithms are presented. The emphasis is on visual diagnostics and interpretation by showing how these work for the examples given. Section 16.1 of this chapter introduces the basic concepts of multivariate regression and of multivariate and hyperspectral images. In Sect. 16.2 the hyperspectral imaging technique used and the two examples (cheese and textile) are explained. Also some sampling issues are discussed here. Principal component analysis (PCA) is a powerful latent variablebased tool for cleaning images. Section 16.3 describes PLS quantitative model building and diagnostics, both numerical and visual for the cheese example, and finishes with PLSDA qualitative modeling for the textile example.
CITATION STYLE
Geladi, P., Grahn, H., & Esbensen, K. H. (2017). Latent variable regression for laboratory hyperspectral images. In Partial Least Squares Path Modeling: Basic Concepts, Methodological Issues and Applications (pp. 339–365). Springer International Publishing. https://doi.org/10.1007/978-3-319-64069-3_16
Mendeley helps you to discover research relevant for your work.