基于支持向量机的激光成像雷达地形重采样

Terrain re-sampling of imaging lidar based on support vector machines

针对机载三维激光成像雷达测量地形高程的特点,提出了一种基于最小平方支持向量机和Shepard的激光成像雷达扫描的地形重采样方法。该方法利用支持向量机是基于结构风险最小化准则,对带有噪声的数据拟合方面具有较好的泛化能力的优点,结合Shepard的局部插值能力,使重建的地形具有局部失真小、总体保持最优的特点。仿真实验结果达到了较高的地形测量精度。

According to the advantage of three-dimensional imaging lidar measuring terrain altitude, a terrain re-sampling method for imaging lidar based on Least Square-Support Vector Machines （LS-SVM） and Shepard is presented. The method has better generalization ability for data fitting with noise by adopting support vector machine method based on the principle of structural risk minimization. Combined with the local interpolation ability of Shepard method, re-sampling terrain has the features of low local distortion and best total effect. The emulation results show the validity and practical value of the method.