神经网络技术在LIDAR测高数据处理中的应用

Study on the Data Processing of LIDAR Height Based on Neural Networks

介绍了先进的激光雷达测量技术,其缺点之一是LIDAR测高数据存在系统偏差。介绍了系统偏差补偿的传统方法思路,如附加系统参数法和最小二乘配置法等。论述了神经网络BP算法的思想及其补偿系统偏差的原理,并列出了BP算法的具体网络模型结构与计算步骤。结合一个具体工程实例,在系统偏差利用神经网络方法补偿之后,LIDAR测高精度有较大提高。最后,得出了一些有益的结论。

At first,the LIDAR（Light Detection and Ranging） technique is introduced.One of the disadvantages of the LIDAR technique is that the LIDAR height has systemic error.Four traditional methods for compensating systemic errors,such as the method of adding systematic parameters and the least-squares collection method are introduced.Then,the BP algorithm of neural network is introduced briefly.A neural network based method for compensating systemic errors is discussed.The structure of BP network,its calculation steps and the principle of this method are introduced in detail.According to one engineering project,the mean standard error of the LIDAR height can be improved much after compensating systemic errors based on neural networks.It is shown that the proposed method based on neural network is good for compensating systemic error.At last,some conclusions are reached.