The Profiles Based on Ridge and Valley Lines to Extract Shoulder Lines on the Loess Plateau

The shoulder line is fundamental to geomorphic evolution and erosion monitoring research on the Loess Plateau, which represents boundary between positive terrain (intergully) negative (inner gully). existing extraction methods mainly suffer problems of unclear geological significance, poor landform application, low efficiency algorithms. This paper proposes a new loess automatic method, in topographic feature points (ridge valley points) were used as endpoints generate continuous profiles, two parameters, analysis operator size (L) filter threshold (σ), created for point from each profile. method can be applied complex landforms such lines terraces extracts. Herein, three typical areas Dongzhi Plateau selected assess performance digital elevation model (DEM) with resolution 5 m was source data. accuracy assessment index Euclidean distance offset percentage (EDOP), original evaluation improved based Structure Motion–Multiview Stereo (SfM-MVS) technology. experimental results showed that average proposed test reached 89.3%, higher than multidirectional hill-shading P-N methods. Via testing different areas, it could concluded less affected by area region, approach exhibited suitable robustness. Simultaneously, optimal values parameters L σ examined. study increases possibility accurate large Plateau.