摘要
为提高机载激光测深(airborne LiDAR bathymetry,ALB)精度,需对ALB原始测深偏差进行修正。通过顾及ALB测深偏差诸影响参数,基于神经网络方法,利用实测样本数据训练获得ALB神经网络偏差模型,然后基于神经网络偏差模型对ALB原始测深数据进行修正。结果表明ALB原始测深偏差较大,需进行修正。传统线性偏差模型和本文神经网络模型均能有效修正ALB测深偏差。相比线性偏差模型,神经网络偏差模型具有更高内符合和外符合精度。基于神经网络的ALB偏差建模方法能有效提高ALB测深精度。
Raw depth derived by airborne LiDAR bathymetry(ALB)needs to be corrected to improve the accuracy of ALB.In this paper,ALB bias model based on neural network is trained by using the measured ALB data considering the various influence parameters,and then the raw ALB depth is corrected by using the neural network bias model.Results show that the raw bias of ALB depth is large and needs to be corrected.Both the traditional linear bias model and the proposed neural network model can effectively correct the ALB depth bias.Compared with the linear bias model,the neural network model has higher accuracy.The ALB bias model based on neural network can effectively improve the accuracy of ALB.
作者
周丰年
许宝华
赵兴磊
ZHOU Fengnian;XU Baohua;ZHAO Xinglei(The Survey Bureau of Hydrology and Water Resources of Yangtze Estuary,Shanghai 200136,China;College of Information Science and Engineering,Shandong Agricultural University,Taian 271018,China)
出处
《海洋测绘》
CSCD
北大核心
2022年第3期5-8,共4页
Hydrographic Surveying and Charting
基金
江苏省自然资源发展专项资金(海洋科技创新)(JSZRHYKJ202002)。
关键词
机载激光测深
测深偏差
神经网络
模型构建
精度评估
airborne LiDAR bathymetry
depth bias
neural network
model building
accuracy assessment
