Abstract
Acquiring accurate and coloured 3D representations of underwater scenes is of prime importance in many applications such as ship hull inspection, offshore structure assessment, archaeology or the detection of hazards such as underwater mines. While video and laser-based 3D reconstruction approaches are limited to very short ranges, the use of SONAR sensors enable the recognition and localisation of man-made structures at centimetre-level. We combine mid-range SONAR mapping with video mapping onto a high-resolution CAD model of the object of interest. We apply this 3D model-based video mapping approach on field data gathered by AUV during lake trials. Through this experiment, we demonstrate the interest in using multi-modal sensing and reference CAD models to obtain high-resolution coloured 3D representations of man-made objects.
Original language | English |
---|---|
Title of host publication | OCEANS 2017 - Anchorage |
Publisher | IEEE |
Pages | 1-5 |
Number of pages | 5 |
ISBN (Electronic) | 9780692946909 |
Publication status | Published - 25 Dec 2017 |
Event | OCEANS 2017 - Anchorage - Anchorage, United States Duration: 18 Sept 2017 → 21 Sept 2017 |
Conference
Conference | OCEANS 2017 - Anchorage |
---|---|
Country/Territory | United States |
City | Anchorage |
Period | 18/09/17 → 21/09/17 |
ASJC Scopus subject areas
- Oceanography
- Automotive Engineering
- Water Science and Technology
- Acoustics and Ultrasonics
- Instrumentation
- Ocean Engineering