［Objective］The aim of this study was to explore the possibility of extracting surface microtopographic features on crop-covered slopes， and realize three-dimensional modeling and dynamic monitoring of microtopography under crop-covered conditions.［Methods］Using the SfM-MVS（Structure from motion with multi-view stereo）method，the maize-covered slope was used as the research object，and the bare slope was used as the control. Based on the markers of bare slope and simulated microtopography，the extraction and accuracy evaluation of surface microtopography under maize-covered conditions were carried out.［Results］1）The errors of the control points on the maize-covered slope were less than 0.002 m，and the number of sparse point clouds and dense point clouds were 2.1 and 2.6 times that of the bare slope，respectively. 2）The results based on C2C（Cloud to Cloud）showed that 85.12% of the distances between maize-covered slope and bare slope point clouds were less than 0.001 m. The results of DOD（DEM of Difference）method showed that 96.07% of the height differences between maize-covered slope and bare slope DEM were less than 0.003 m，and the overall accuracy reached millimeter level. 3）The microtopographic extraction of bare slope markers were closer to the actual values，and the accuracy of length，width，and depth of markers in all directions was above 98%. Although the accuracy of the lateral markers on the maize-covered slope decreased due to the superposition of vegetation，the accuracy was still above 97%.［Conclusion］The measurement method based on SfM-MVS can be used as a high-accuracy measurement method to extract the microtopographic fluctuation characteristics of maize-covered slopes.