测绘学报 ›› 2023, Vol. 52 ›› Issue (6): 944-955.doi: 10.11947/j.AGCS.2023.20210603

• 摄影测量学与遥感 • 上一篇    下一篇

机载激光测深近水面渗透误差修正的半经验波形分解方法

王丹菂, 徐青, 邢帅, 林雨准, 张国平   

  1. 信息工程大学地理空间信息学院, 河南 郑州 450052
  • 收稿日期:2021-10-28 修回日期:2022-05-16 发布日期:2023-07-08
  • 通讯作者: 邢帅 E-mail:xing972403@163.com
  • 作者简介:王丹菂(1993-),女,博士,计师,研究方向为机载激光测深技术。E-mail:wdd_93@163.com
  • 基金资助:
    国家自然科学基金(41876105; 41371436)

Semi-empirical waveform decomposition method for correction of near water surface penetration error in airborne LiDAR bathymetry

WANG Dandi, XU Qing, XING Shuai, LIN Yuzhun, ZHANG Guoping   

  1. Institute of Geospatial Information, Information Engineering University, Zhengzhou 450052, China
  • Received:2021-10-28 Revised:2022-05-16 Published:2023-07-08
  • Supported by:
    The National Natural Science Foundation of China (Nos. 41876105; 41371436)

摘要: 信号检测精度是影响机载激光测深最终测量成果的一项关键因素。针对绿激光在水面存在的近水面渗透现象,为提高水面信号检测精度,本文提出一种半经验波形分解方法。该方法通过简化激光辐射传输过程构建一种符合波形实际情况的半经验信号卷积模型,并利用基于航迹和影像数据人工选取的深水波形样本估计模型中水体参数初值及取值范围,在波形先验的约束下基于信赖域算法实现波形各组成部分的精确重构,从而确定水面信号位置并修正近水面渗透误差。试验结果表明,本文方法在波形分解中将理论与经验结合,可较好地适应不同水深下的波形,能够在保证高拟合度的同时提升水面信号的检测精度,相比去卷积算法和传统波形分解方法,本文方法在精度上分别提高了44%和51%。

关键词: 机载激光测深, 波形分解, 卷积, 水体衰减系数, 信号检测

Abstract: The accuracy of signal detection is a key factor that affects the final measurement results of airborne LiDAR bathymetry. To solve the problem that green laser penetrates the water column in near water surface and improve the accuracy of the detected water surface signal, a semi-empirical waveform decomposition method is proposed. A semi-empirical signal convolution model that conforms to the field waveforms is constructed by simplifying the laser radiation transmission model. Deep water waveform samples are manually collected using the flight trajectory and image to estimate the initial values and ranges of the water column parameters in the model. Based on the trust region algorithm, each component of the waveform is precisely reconstructed with the constraints of the waveform priors, so the position of the water surface signal is obtained, and the near water surface penetration error is corrected. The experimental results show that the proposed method combines theory and experience in waveform decomposition, adapting to waveforms with different water depths and improving the accuracy of the detected water surface signal with good waveform fitting. Compared with the deconvolution algorithm and the traditional waveform decomposition method, the accuracy of the proposed method achieves 44% and 51% improvement, respectively.

Key words: airborne LiDAR bathymetry, waveform decomposition, convolution, water attenuation coefficient, signal detection

中图分类号: