A Terrain Model Simplification Method Based on Adaptive Threshold Selection

This paper presents a fast simplification strategy for the terrain model based on identifying the terrain features. In this strategy, the original data is organized on a hierarchical structure with multi-resolution, which reduced effectively the range of searching feature points. Thus, the feature points can be selected fast according to our rules. Taking the feature points as the centers, we can obtain the feature areas by using the proposed division algorithm. The aim is to provide the criterion for the error evaluation and the threshold enactment. Based on the feature areas, we proposed the multi-threshold algorithm, which improved the precision and the self-adaptability of the result model greatly. Finally, many experiments have been conducted on the various benchmark data, which shows that our method is very efficient and the result model can preserve the detailed features with a quite high precision even as the simplification ratio is very high.