SSA-BP神经网络在无人机点云孔洞修补的应用  

Application of SSA-BP neural network in UAV point cloud hole repair

作　　者：吕富强 唐诗华 张炎 宋晓辉 胡鹏程 李翥 LV Fuqiang;TANG Shihua;ZHANG Yan;SONG Xiaohui;HU Pengcheng;LI Zhu(College of Geomatics and Geoinformation,Guilin University of Technology,Guilin 541004,China;Guangxi Key Laboratory of Spatial Information and Geomatics,Guilin 541004,China)

机构地区：[1]桂林理工大学测绘地理信息学院,广西桂林541004 [2]广西空间信息与测绘重点实验室,广西桂林541004

出　　处：《测绘通报》2023年第5期130-134,共5页Bulletin of Surveying and Mapping

基　　金：国家自然科学基金(42064003)。

摘　　要：为了解决无人机点云数据中的孔洞修补问题,本文提出了基于麻雀搜索算法(SSA)优化BP神经网络的无人机点云孔洞修补方法。首先利用麻雀搜索算法对传统的BP神经网络进行初始权重与阈值的优化,再将麻雀搜索算法优化后的BP神经网络算法(SSA-BP)运用于无人机点云数据中孔洞的修补。为了验证算法的可行性,将SSA-BP神经网络与传统的BP神经网络、最小二乘支持向量机(LSSVM)两组算法进行精度比较。试验结果表明:SSA-BP神经网络算法的修补精度高于另外两组对比算法,且SSA优化后的BP神经网络稳定性更强,在复杂地形孔洞的修补中仍具有较好的修补效果。In order to solve the problem of hole repair in UAV point cloud data,a back-propagation neural networkhole repair method was proposed based on sparrow search algorithm(SSA).The sparrow search algorithm was used to optimize the initial weight and threshold of the traditional BP neural network,and then the BP neural network algorithm(SSA-BP)optimized by the sparrow search algorithm was applied to repair the holes in uav point cloud data.In order to verify the feasibility of the algorithm,the accuracy of SSA-BP neural network was compared with that of traditional BP neural network and least square support vector machine(LSSVM)algorithms.The experimental results show that the repair accuracy of SSA-BP neural network algorithm is higher than the other two groups of comparison algorithms,and the SSA-BP neural network is more stable,and it still has a good repair effect in the repair of complex terrain holes.

关 键 词：孔洞修补 麻雀搜索算法 优化 BP神经网络 精度比较 

分 类 号：P237[天文地球—摄影测量与遥感]