For the industrial application scenario of waste copper alloy recycling and classification, two machine learning algorithms based on microjoule high-frequency laser-induced breakdown spectroscopy (MH-LIBS) combined with artificial neural network (ANN) and support vector machine (SVM) are used. Seven copper alloy samples (H59, H62, H70, H85, H96, HPb59-1, HPb62) collected in point and motion modes were classified and recognized, respectively. The results show that ANN and SVM can achieve 100% accuracy in classifying the copper alloys collected in point mode. The classification accuracy for the copper alloys collected in motion mode is 100% and 99.86%, respectively. It can be seen that the microfocus high-frequency laser-induced breakdown spectroscopy system combined with machine learning algorithms can realize the fine classification of copper alloys, which is suitable for the rapid analysis of waste copper alloys on site.