Chapter 7Application Challenges of Underwater Vision

Nuno Gracias1, Rafael Garcia1, Ricard Campos1, Natalia Hurtos1, Ricard Prados1, ASM Shihavuddin2, Tudor Nicosevici1, Armagan Elibol3, Laszlo Neumann1 and Javier Escartin4

1Computer Vision and Robotics Institute, University of Girona, Girona, Spain

2École Normale Supérieure, Paris, France

3Department of Mathematical Engineering, Yildiz Technical University, Istanbul, Turkey

4Institute of Physics of Paris Globe, The National Centre for Scientific Research, Paris, France

7.1 Introduction

Underwater vehicles, either remotely operated or autonomous, have enabled a growing range of applications over the last two decades. Imaging data acquired by underwater vehicles have seen multiple applications in the context of archeology (Eustice et al. 2006a), geology (Escartin et al. 2009; Zhu et al. 2005), or biology (Pizarro and Singh 2003) and have become essential in tasks such as shipwreck inspection (Drap et al. 2008), ecological studies (Jerosch et al. 2007; Lirman et al. 2007), environmental damage assessment (Gleason et al. 2007a; Lirman et al. 2010), or detection of temporal changes (Delaunoy et al. 2008), among others. Despite their acquisition constraints, which often require the underwater vehicle to navigate at a close distance to the seafloor or the structure of interest, imaging sensors have the advantage of purveying higher resolution and lower noise when compared with the traditional sensors for seafloor surveying such as ...

Get Computer Vision in Vehicle Technology now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.