Skip to main content
. 2018 May 9;410(18):4541–4554. doi: 10.1007/s00216-018-1111-x

Fig. 3.

Fig. 3

Classification of basalis and functionalis regions by spectral analysis using PCA-LDA, SPA-LDA and GA-LDA on FPA-FTIR derived data (red = functionalis, blue = basalis): All specimens. a Cost/function plot identifying the optimal number of PCs to be used for PCA. b Scores plot graphically representing classification by PCA-LDA. The x-axis represents the sample index and the y-axis the discriminant factor 1 (DF1). c Wavenumber selection for SPA-LDA. d Scores plot graphically representing classification by SPA-LDA. The x-axis represents the sample index and the y-axis DF1. e Wavenumber selection for GA-LDA. f Scores plot graphically representing classification by GA-LDA. The x-axis represents sample index and the y-axis DF1