In order to solve the problem of uncertain or incorrect classification due to local feature confusion in land‐use classification based on high‐resolution remote sensing images, a land‐use‐information‐extraction method based on feature selection was proposed. Firstly, multi‐scale segmentation was carried out with existed land‐use data, and the optimal segmentation scales about different land features were obtained. Then, based on the prior knowledge from attribute information of land‐use data, the feature selection was carried out. Finally, the image classification results were obtained by fuzzy classification calculation. In order to evaluate the classification accuracy, the method based on feature selection was compared with random forest and decision tree algorithm. The results showed that classification accuracy of the method adopted in this paper were higher.