Uncertainty-Aware Quality Control of Airborne Bathymetric LiDAR Point Clouds in Tidal Flats: A Data-Driven Approach
IEEE Geoscience and Remote Sensing Letters · 2026 · 23 · 1-5
DOI: 10.1109/lgrs.2026.3673075 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Geoscience and Remote Sensing Letters, 2026. Verfasst von Ahram Song. Die Arbeit nennt 21 Literaturangaben. Gefördert von: Korea Institute of Marine Science and Technology Funded by the Ministry of Oceans and Fisheries.
Abstract (Originalsprache)
Airborne bathymetric LiDAR (ALB) is widely used to map intertidal and shallow-water zones; however, reliable terrain reconstruction is hindered by sea-surface returns and depth-dependent uncertainty, especially in regions where ground truth (GT) is sparse. This letter introduces an uncertainty-guided quality-control (QC) strategy designed to evaluate and filter ALB points prior to surface modeling. Pointwise semantic classification estimates sea-surface probabilities, while Monte Carlo dropout quantifies epistemic uncertainty, which is combined into a confidence score for probabilistic filtering. Ground-truth-free indicators including spatial completeness, surface roughness, and depth-dependent vertical stability are then computed from the filtered surfaces. A knee-point criterion is applied to determine a threshold that balances coverage preservation and surface stability. Experiments across eight tidal-flat subregions show that the proposed strategy reduces surface roughness while maintaining spatial completeness and minimal elevation bias. Removed grid cells occur mainly in deeper tidal channels where uncertainty levels are highest, indicating that the method selectively suppresses unreliable measurements. Overall, the results demonstrate that uncertainty-guided filtering can support practical QC decisions in ALB surveys without relying on dense reference bathymetry.
Förderung
- Korea Institute of Marine Science and Technology Funded by the Ministry of Oceans and Fisheries
Zitieren
APA
Song, A. (2026). Uncertainty-Aware Quality Control of Airborne Bathymetric LiDAR Point Clouds in Tidal Flats: A Data-Driven Approach. IEEE Geoscience and Remote Sensing Letters, 23, 1-5. https://doi.org/10.1109/lgrs.2026.3673075
Vancouver
Song A. Uncertainty-Aware Quality Control of Airborne Bathymetric LiDAR Point Clouds in Tidal Flats: A Data-Driven Approach. IEEE Geoscience and Remote Sensing Letters. 2026;23:1-5. doi:10.1109/lgrs.2026.3673075
BibTeX
@article{Song2026,
author = {Song, Ahram},
title = {Uncertainty-Aware Quality Control of Airborne Bathymetric LiDAR Point Clouds in Tidal Flats: A Data-Driven Approach},
journal = {IEEE Geoscience and Remote Sensing Letters},
year = {2026},
volume = {23},
pages = {1--5},
doi = {10.1109/lgrs.2026.3673075}
}
EUVIMED