大面积带状区域高程拟合方法研究

摘要

This article described the necessity of research on elevation fitting methods in the large and zonal area , and presented the LS-SVM,Back Propagation neural network ,and quadratic polynomial curve surface fitting method . With engineering examples are carried out ,focusing on three fitting methods in elevation with a large and zonal area .The results show the LS-SVM obtained a higher elevation fitting comply with internal and external precision in the large and zonal area ,and the method has a more good plasticity and generalization ability .%介绍大面积带状区域高程拟合方法研究的必要性，并列举了最小二乘支持向量机、BP神经网络、二次曲面拟合三种高程拟合方法。结合工程实例，重点对三种拟合方法在同一个大面积带状区域的高程拟合结果进行对比分析。结果表明，针对本文大面积带状区域，最小二乘支持向量机进行高程拟合获得较高的内外符合精度，该方法具有较好的可塑性和更强的泛化能力。