面向地形数据的点云简化算法

针对地形点云数据量大、表面特征复杂多样等特点,提出面向地形数据的点云简化算法。基于K-D Tree搜索各点K邻域,构建点集空间拓扑关系;应用移动最小二乘法计算各点曲率,通过曲率的划分,在平缓区域按距离进行简化,保证整个算法的效率;在突变区域根据曲率简化,确保曲率变化大的关键特征信息不丢失,从而实现点云数据的简化。利用基于熵理论的定量评价方法,通过实例验证该方法的可行性和普适性。

Point Cloud Simplification Algorithm for Terrain Data

Large amounts of data will have an adversely impact on point cloud’s processing, storage, transmission and display. In order to improve the point cloud’s data processing efficiency, it is necessary to simply point cloud data and reduce data redundancy. Due to the large quantity and complex surface features of terrain point cloud data, the point cloud simplification algorithm for terrain data is proposed.We use K-D Tree to search the point’s K neighborhood and the moving least squares method to calculate the curvature of each point. Through the vision of curvature, point cloud data is simplified based on distance in the flat area to improve the efficiency of the algorithm and on the basis of curvature in the steep area to retain the feature information. According to the quantitative evaluation method,based on the entropy theory and verified by the example, it is indicated that the proposed new simplification algorithm for terrain data can simply the data in high-fidelity with high-precision. It is highly efficient feasible and universally applicable.