A heuristic DEM generalization by combining catchments

Abstract

Multi-scale digital elevation model (DEM) is an important content of digital terrain analysis. To generate multi-scale DEM, this study reports a heuristic DEM generalization method. The key of DEM generalization is to maintain the main topographic features and remove the minor topographic features. Our method takes catchment as generalizing object, not single grid pixel as generalizing object. In this way, DEM simplification can be achieved on the basis of maintaining the main terrain characteristics. Compared with the resample method, VIP method and low pass filter method, the proposed method can preserve the main topographic features well when the minor topographic features are deleted. Our main contribution lies in: taking terrain information rather than raw data as the generalizing object, which improves the rationality of the generalizing results; and the heuristic method proposed in this paper has good operability and high automation. Highlights A heuristic DEM generalization method by combining catchments is developed; The proposed DEM simplification can maintain the main terrain characteristics; The heuristic method proposed in this paper has good operability and high automation;