학술논문

Statistical classification methodology of SHOALS 3000 backscatter to mapping coastal benthic habitats
Document Type
Conference
Source
2007 IEEE International Geoscience and Remote Sensing Symposium Geoscience and Remote Sensing Symposium, 2007. IGARSS 2007. IEEE International. :3178-3181 Jul, 2007
Subject
Geoscience
Signal Processing and Analysis
Backscatter
Sea measurements
Sea surface
Surface emitting lasers
Laser radar
Surface topography
Laser modes
Pulse measurements
Surface waves
Optical surface waves
bathymetric LiDAR
waveform
multivariate analysis
habitat classification
Language
ISSN
2153-6996
2153-7003
Abstract
The Scanning Hydrographic Operational Airborne LiDAR Survey (SHOALS) consists of a bathymetric LiDAR system which provides high precision measurements of water depth. Even though the acquisition is focused on depth accuracy, the return signal, i.e. waveform, contains other relevant information because of integration signatures from the water surface, the water column and the sea-bed. This paper highlights the benthic characterization in extracting statistical parameters derived from the bottom backscatter. In applying multivariate analysis (K-means), it is significantly proven that signals derived from habitat, described as statistically homogeneous throughout ground-truth analysis, are (1) similar within an intra-habitat view, while they are (2) different between themselves.