Temperature-perturbed two-dimensional generalized correlation characteristic slice spectra combined with multivariate method to identify adulterated milk

•The discrimination method for adulterated milk was proposed using temperature-perturbed 2D correlation characteristic slice spectra.•2D correlation slice spectra were used to extract important feature information of different brands of milk and adulterants.•2D correlation characteristic slice spectral models are superior to 2D correlation, and 1D spectral models.•2D correlation characteristic slice spectra greatly compressed 2D correlation spectra with a compression rate of 98.4%.