Based on ecognition software platform, the method of object-oriented computer automatic classification combined with manual visual interpretation was used to extract the land use data of the study area year by year. Finally, three methods are used to verify the accuracy of the data: field sample point survey, high-resolution image recognition and Google Earth sample point recognition.

Remote sensing images are preprocessed by radiation correction, ortho rectification, fusion and mosaic. The actual surface area corresponding to the minimum spot area is not less than 0.1 h ^, the polygons have no overlap or gap, and the spot attributes have no vacancy or redundancy. Before remote sensing image interpretation, the land use remote sensing interpretation marks are established by using remote sensing image, typical survey and field comparison. Based on remote sensing images, combined with interpretation marks, land use types are extracted. Re examination of interpretation results: no less than 5% of the total map spots shall be selected for verification. The number and results of field verification samples meet the requirements of technical specification for remote sensing monitoring of soil and water conservation (sl592-2012). For verification spots, 10% of them are selected as verification samples for field verification.