Abstract: An automatic extraction method of expressway road surface from mobile laser scanning(MLS)data is presented.Through an analysis of the spatial characteristics of expressway MLS pointcloud data,an approximate plane constraint(APC)method,an ordering least square method of slopeestimation and multi-scale window iterative analysis(MWIA)method were developed to extract initialground seed points on the grid,and then all road surface points are extracted from the original pointsbased on local slope.Finally,two groups of actual point cloud data were selected for an experiment,the results show that the proposed method can quickly and accurately extract a road surface from MLSdata.The correctness of our road extraction results was 95.74%and completeness was 98.11%.