The harvesting period of spring tea significantly affects its economic value and consumer preference. To quickly identify different harvesting periods of spring tea, we employed laser-induced breakdown spectroscopy (LIBS) combined with machine learning algorithms. This approach was used to identify the before-brightness tea and before-rain tea of Mt. Lushan fog tea and Dog bull head tea. One hundred spectra were collected for each type of tea leaves and tea infusion, and the training and test sets were randomly divided in a ratio of 3∶2. The LIBS spectra were pre-processed with baseline correction and then 11 sets of spectral data were preferentially selected, and input into the linear discriminant analysis (LDA), support vector machines (SVM), K-nearest neighbor (KNN) and ensemble machine learning (EML) classification models for analysis, respectively. Findings showed that combining tea leaves and tea infusion data effectively identified the spring tea's harvesting period. This fusion approach exhibited superior stability and robustness. Specifically, the LDA model achieved recognition rates of 98.60% and 99.38% in the test sets for Mt. Lushan fog tea and Dog bull head tea, respectively. Therefore, this study demonstrates the feasibility of integrating LIBS with machine learning algorithms to discern different harvesting periods of spring tea.