In this study, ultra-high temperature (UHT) milk and reconstituted milk were discriminated by high resolution quadrupole-time of flight mass spectrometry (LC-Q-TOF-MS) combined with chemometrics based on non-targeted metabolomic approaches. Sample pretreatment was performed prior to chromatographic separation on a C18 column, and the mass spectrometer was operated in the full scan mode. The obtained data were preprocessed before being imported into SIMCA-P 14.1 for principal component analysis (PCA) and partial least squares-discrimination analysis (PLS-DA). A total of 14 characterization factors were selected in both positive and negative ion modes to establish a discriminant model. The results showed that this model could be used as a robust approach to discriminate between UHT and reconstituted milk and provide a theoretical basis for the identification of reconstituted milk.