Underwater Photogrammetry and Spectral Imaging for Aquatic Mapping

Scope and topics

This Special Session focuses on the mapping and monitoring of submerged environments using underwater
photogrammetry and multispectral or hyperspectral imaging. The common objective is to obtain reliable
geometric and thematic information about submerged surfaces, habitats and objects. Spectral imaging is
considered specifically in relation to aquatic mapping, rather than as a separate theme covering unrelated
terrestrial applications.
We welcome underwater photogrammetric reconstruction, aquatic spectral mapping and studies combining
the two. Contributions need not use both approaches in the same workflow. Integration may involve
registering spectral observations with photogrammetric 3D models, combining structural and spectral
descriptors for habitat classification, or comparing complementary observations of the same site.
Methodological topics include geometric, radiometric and spectral calibration; modelling refraction at
camera ports and the air-water interface, as applicable; wavelength-dependent attenuation and scattering;
water-column correction; image orientation and 3D reconstruction; sensor synchronisation; and co-
registration of imagery, spectral products and surface models. Contributions should distinguish
improvements in visual appearance from preservation of quantitative geometric or spectral information.
Acquisitions may use diver-operated cameras, remotely operated vehicles (ROVs), autonomous underwater
vehicles (AUVs), or uncrewed marine vessels. UAV, crewed-aircraft and satellite observations are relevant
when they support through-water mapping or provide complementary aquatic information. Above-water
retrievals of submerged features must account for water depth and optical conditions. GNSS may support
surface-platform positioning, while underwater navigation and georeferencing may use acoustic, inertial or
visual methods and surveyed control.
Geospatial Artificial Intelligence (GeoAI) contributions may address habitat classification, object
recognition, segmentation of georeferenced imagery or 3D models, and change detection. Particular interest
lies in combining spatial structure with spectral information. Studies should explain how reference-data
quality, spatial sampling, uncertainty and validation across sites or survey dates affect reliability. Image
restoration methods should assess whether features used for mapping or spectral interpretation are preserved.
Expected applications include benthic habitat mapping, underwater archaeology, inspection of submerged
infrastructure, shallow-water bathymetry and monitoring of aquatic vegetation or seabed change. Examples
include RGB photogrammetric surveys of submerged structures, multispectral or hyperspectral mapping of
benthic habitats, and habitat maps combining spectral classes with 3D surface geometry. These examples
define complementary contribution types, rather than a requirement to combine every sensor or platform.
Methodological papers, field studies, comparative experiments and critical assessments of unsuccessful
configurations are welcome. Contributions should document acquisition conditions, processing choices and
validation appropriate to their geometric or thematic objectives. Open datasets, benchmark sites and
repeatable field protocols are encouraged. The session aims to establish when geometric and spectral
approaches can be used independently or together to produce dependable aquatic maps.

List of topics:

Organizers