 This study compared the performance of four machine learning classifiers, support vector machine, SVM, normal bays, NB, classification and regression tree, CART, and K nearest neighbor, K&N, on very high resolution images. The results show that SVM and NB were superior to CART and K&N, achieving high classification accuracy, greater than 90%. Additionally, the optimal settings of tuning parameters varied significantly depending on the size of the training set. Furthermore, the size of the training set had a significant impact on the classification accuracy, with smaller sets resulting in lower accuracy. Finally, this research suggests that it is important to select the right classifier and adjust the appropriate parameters for different data sets. This article was authored by Yu Guoqin, Weichi Zhou, Jing Liyan, and others.