Figure 4: Principal component analysis of the hyperspectral CARS sample image. (a) Cumulative variance plot shows that only two principal components (PCs) are needed to cover ~90% of the variance in the data set. No appreciable increase in the cumulative variance is observed after PC2 as shown for PC3 to PC6. (b) Score plot of the first two principal components (PC1 and PC2). Each point on the plot corresponds to one pixel in the image. (c) The color assignments on the retrieved PCA image depend on which quadrant the pixel object is located on the score plot in (b). Scale Bar: 30 μm.