Poster + Presentation + Paper
17 October 2023 Coastal water feature extraction using airborne hyperspectral imagery in shallow estuarine water
Author Affiliations +
Conference Poster
Abstract
Space Coast Florida satellite and airborne multispectral and hyperspectral imagery hold the potential for discrimination of bottom features in water bodies. Automated bottom feature extraction techniques range from Principal Component Analysis (PCA), Discriminant Analysis and Canonical Correlation as well as the well know Minimum Noise Fraction (MNF) that is related to PCA. Application of these multivariate methods can be performed to reduce the dimensionality of the imagery (bands) as well as spatial (pixel) noise reduction and spectral noise reduction. The techniques can also help reduce spectral band intercorrelation. The MNF methods can thus be used for band selections or channels that contain known geophysical phenomena, such as “bottom type features” when used with ground validation or field sampling methods. Following application of the above noise reduction methods the selected bands can be used with feature analysis methods to perform training of neural network or artificial intelligent algorithms using selected pixel (raster data) in order to perform the creation of vector-based areas or regions based upon spectral shape similarity as demonstrated to select submerged water feature areas.
Conference Presentation
(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Charles R. Bostater Jr. and T. Ghir "Coastal water feature extraction using airborne hyperspectral imagery in shallow estuarine water", Proc. SPIE 12728, Remote Sensing of the Ocean, Sea Ice, Coastal Waters, and Large Water Regions 2023, 1272809 (17 October 2023); https://doi.org/10.1117/12.2680314
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Reflectivity

Feature extraction

Hyperspectral imaging

Remote sensing

Vegetation

Water

Raster graphics

Back to Top