Real-Time Monocular Visual Odometry for Turbid and Dynamic Underwater Environments

Maxime Ferrera 1, 2 Julien Moras 1 Pauline Trouvé-Peloux 1 Vincent Creuze 2
2 ICAR - Image & Interaction
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
Abstract : In the context of underwater robotics, the visual degradation induced by the medium properties make difficult the exclusive use of cameras for localization purpose. Hence, many underwater localization methods are based on expensive navigation sensors associated with acoustic positioning. On the other hand, pure visual localization methods have shown great potential in underwater localization but the challenging conditions, such as the presence of turbidity and dynamism, remain complex to tackle. In this paper, we propose a new visual odometry method designed to be robust to these visual perturbations. The proposed algorithm has been assessed on both simulated and real underwater datasets and outperforms state-of-the-art terrestrial visual SLAM methods under many of the most challenging conditions. The main application of this work is the localization of Remotely Operated Vehicles used for underwater archaeological missions, but the developed system can be used in any other applications as long as visual information is available.
Complete list of metadatas

Cited literature [69 references]  Display  Hide  Download

https://hal-lirmm.ccsd.cnrs.fr/lirmm-02012078
Contributor : Vincent Creuze <>
Submitted on : Friday, February 8, 2019 - 2:16:54 PM
Last modification on : Tuesday, March 26, 2019 - 2:12:02 PM
Long-term archiving on: Thursday, May 9, 2019 - 2:22:47 PM

File

sensors-19-00687.pdf
Publication funded by an institution

Identifiers

Citation

Maxime Ferrera, Julien Moras, Pauline Trouvé-Peloux, Vincent Creuze. Real-Time Monocular Visual Odometry for Turbid and Dynamic Underwater Environments. Sensors, MDPI, 2019, Special Issue: Intelligent Underwater Systems: Sensing, Communication, Networking and Applications, 19 (3), pp.687-709. ⟨10.3390/s19030687⟩. ⟨lirmm-02012078⟩

Share

Metrics

Record views

66

Files downloads

58