Title: Benchmarking underwater photogrammetry

Target Audience: PhD students, researchers, and practitioners

Level: Beginner to intermediate

Date and time: June 10, 2024 at 1pm

Duration: 4 hours

Teaching resources will be distributed (i.e. the PowerPoint presentations, image datasets, python codes).

The underwater environment represents one of the modern frontiers and opportunities for scientific and technological advances. While ocean exploration is recognized as a priority for humankind (sustainable) development, underwater navigation, mapping and monitoring still present unique challenges. One of the key factors is the complexity of acquiring stable and reliable reference measurements against which to verify and validate results produced using geomatic techniques such as photogrammetry. This is also reflected in the scarcity of publicly available benchmark datasets, an essential tool in scientific research as a method of comparison, validation and improvement of theories and techniques. This tutorial will provide the audience with an introduction to the crucial aspects of benchmarking underwater photogrammetry, both in terms of geometric accuracy and color reproduction. We will present an overview of the techniques used to measure underwater control points, perform direct georeferencing underwater and assess color correction/enhancement/restoration methods. Topics covered will include definition and materialization of reference systems underwater, accuracy assessment and method of controls with scale bars, underwater geodetic measurements, acoustic positioning systems, pressure and inertial sensors, laser based optical scaling, quality metrics for color correction/enhancement/restoration approaches.

The issue of underwater benchmark datasets dissemination and sharing will be addressed in the light of the results of the ISPRS SI2023 project NAUTILUS - uNder And throUgh waTer datasets for geospatIaL studies. The topics will be addressed from a theoretical point of view and several case studies will be analyzed together. Participants are also encouraged to bring their own datasets, showcase their applications, and discuss issues faced in their practices.