Deep Learning Method Suitable for Airborne Laser Bathymetry of Different Water Qualities

Yifan Huang; Yan He; Xiaolei Zhu; Guangxiu Xu

doi:10.3788/CJL241064

Chinese Journal of Lasers, Volume. 52, Issue 1, 0110003(2025)

Deep Learning Method Suitable for Airborne Laser Bathymetry of Different Water Qualities

Yifan Huang^1,2, Yan He^1,2、*, Xiaolei Zhu^1,2, and Guangxiu Xu³

¹Laboratory of Space Laser Information Transmission and Detection Technology, Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Shanghai 201800, China

²Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences, Beijing 100049, China

³Naval Research Institute, Tianjin 300061, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(9)

Fig. 1. Examples of waveforms with different lidar attenuation coefficients (points are seafloor positions)

Download full size

Fig. 2. Removal process of scattering part of water body.(a) Water scattering waveform and waveform after scattering removal; (b) waveforms with different lidar attenuation coefficients after removing water scattering

Download full size

Fig. 3. Sample example.(a) Scanning trajectory of airborne bathymetric lidar; (b) point cloud calculated by using continuous scanning waveform in box of Fig. 3 (a); (c) image of point cloud projection in Fig. 3 (b); (d) corresponding category for each pixel in Fig. 3(c)

Download full size

Fig. 4. Neural network structure

Download full size

Fig. 5. Seafloor points extracted by different methods in different regions. (a) Dacheng Wan; (b) Dazhou Island; (3) Beijiao

Download full size

Fig. 6. Measurement results of Dazhou Island. (a) Water depth distribution of Dazhou Island; (b) bathymetric accuracy of model prediction method on Dazhou Island

Download full size

Table 1. Accuracies of seafloor echo extraction for different training sets
View table
Table 1. Accuracies of seafloor echo extraction for different training sets
Training set $K_{l i d a r} = 0.1$ $K_{l i d a r} = 0.3$ $K_{l i d a r} = 0.5$
Training set A 0.1423 0.6981 0.8017
Training set B 0.8542 0.8858 0.9022

Table 2. Accuracies of seafloor echo extraction with different processing methods
View table
Table 2. Accuracies of seafloor echo extraction with different processing methods
Method Dazhou Island Dacheng Wan Beijiao
Model prediction 0.9217 0.5702 0.2945
Waveform processing 0.5438 0.1034 0.3228

Table 3. Maximum probed water depths for different treatment methods
View table
Table 3. Maximum probed water depths for different treatment methods
Method Dazhou Island Dacheng Wan Beijiao
Model prediction -18.9721 -8.8305 -30.3501
Waveform processing -15.7020 -7.2344 -24.3841

Tools

Get Citation

Copy Citation Text

Yifan Huang, Yan He, Xiaolei Zhu, Guangxiu Xu. Deep Learning Method Suitable for Airborne Laser Bathymetry of Different Water Qualities[J]. Chinese Journal of Lasers, 2025, 52(1): 0110003

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites