Machine learning has achieved great success in many areas today, but the predictive effect of machine learning often depends on the specific problem. An ensemble learning predicts results by integrating multiple base classifiers. Therefore, its ability to adapt to various scenarios is strong, and the classification accuracy is high. In response to the low classification accuracy of darkest source magnitude sets in star/galaxy in the Sloan Digital Sky Survey (SDSS), a star/galaxy classification algorithm based on the Stacking ensemble learning is proposed in this paper. The complete photometric data set is obtained from SDSS-Data Release (DR) 7 and divided into bright source magnitude set, dark source magnitude set, and darkest source magnitude set according to the magnitude. Firstly, the ten-fold nested cross-validation method is used for the darkest source magnitude set, and then the Support Vector Machine (SVM), Random Forest (RF), eXtreme Gradient Boosting (XGBoost) algorithms are used to establish the base-classifier model; the Gradient Boosting Decision Tree (GBDT) is used as the meta-classifier model. Finally, based on galaxies' classification accuracy and other indicators, the classification results are compared with the models of Function Tree (FT), SVM, RF, GBDT, Stacked Denoising Autoencoders (SDAE), Deep Belief Nets (DBN), and Deep Perception Decision Tree (DPDT) etc., and then analyzed. The experimental results show that, the Stacking ensemble learning model improves the classification accuracy of galaxies in the darkest source classification by nearly 10% compared to the function tree algorithm. Compared with other traditional machine learning algorithms, strong lifting algorithms and deep learning algorithms, the Stacking ensemble learning model also has different degrees of improvement.