Fig 7. Investigating the effect of background correction on the accuracy of machine learning-based hyperspectral imaging classification.

Classification accuracy of three machine learning methods (k-means clustering, support vector machine and convolutional neural network) in absorbance (a) and reflectance (b) hypercubes obtained by GT, SB and RB methods. (c—f) Representative images of classification results indicated by c–f in (a) and (b). Scale bar: 100 pixels.