Analysis of the selection of MS1000 sonar image registration parameters for the purpose of automatic detection of moving objects

Natalia Wawrzyniak
Maritime University of Szczecin
Faculty of Navigation
Institute of Geoinformatics
Poland

Grzegorz Zaniewicz
Maritime University of Szczecin
Faculty of Navigation
Institute of Geoinformatics
Poland

Abstract

Stationary high-frequency scanning MSISs (Mechanically Scanned Imaging Sonar) are mainly used to visualize previously known underwater structures and to aid ROVs (Remotely Operated Vehicles) and divers to navigate during underwater inspections and surveys. Their high frequency and ability to draw sonar beam in close to real-time mode allows to track objects situated in their scanning range.
ROVs usually play an additional role in visual inspections of underwater structures and sought objects. Equipped with several propellers, ROVs are able to move in any direction specified by its operator. Steering is also supported by basic navigational sensors as compass and depth sensors. Additionally, thanks to an embedded video camera with LED lightning, real-time image can be instantly sent to controller’s console on the shore.
The proposed approach allows automation of the process of first localization of the moving object (ROV) in the sonar image, provided that a proper selection of parameters for recording MSIS images is made. In the studies, several tests were conducted of the actual influence under real conditions of sonar signal gain factor, TVG correction and scan speed on the detection process. Different tracking scenarios were run for selected settings. Testing algorithm for object localization was implemented in Matlab environment.
Proper selection of recording parameters facilitates separation of the moving object on sonar image. This improves the process of tracking and tracing the moving objects. The study used MS1000 scanning sonar and VideoRay underwater robot explorer to conduct all tests.

Keywords:

scanning sonar; ROV; tracking; underwater navigation

Full Text:

PDF (Polish)

References

Chantler M.J., Lane D.M., Dai D., Williams N., 1996: Detection and tracking of returns in sector-scan sonar image sequences, Radar, Sonar and Navigation vol. 143, no. 3.

Donghwa L., Gonyop K., Donghoon K., Hyun M., Hyun-Taek C., 2012: Vision-based object detection and tracking for autonomous navigation of underwater robots, Ocean Engineering vol. 48, July 2012: 59-68.

Duda J., Ratuszniak N., 2010: Wpływ biegunowego procesu rejestracji na interpretację obrazów z sonaru skanującego, Mapy i zobrazowania powierzchni. Polski Internetowy Informator Geodezyjny, www.Geodezja.pl, Gdańsk.

Kongsberg Mesotech Ltd., 2008a: MS1000 operational manual, Kanada.

Kongsberg Mesotech Ltd., 2008b: SMB File Format, Kanada.

Lane D.M., Chantler M.J., Dai D., 1998: Robust Tracking of Multiple Objects in Sector-Scan Sonar Image Sequences Using Optical Flow Motion Estimation, IEEE Journal of Oceanic Engineering vol. 23.

Lekkerkerk H-J., Theijs M.J., 2011: Handbook of offshore surveying, Skilltrade.

Mazel Ch., 1985: Side Scan Sonar Record Interpretation, Klein Associates Inc.

Ratuszniak N., Pałczyński M., 2010: Method of Visualization for Scanning Sonar Image, Pomiary Automatyka Kontrolna vol.56 nr 12.

Society for Underwater Technology, 2007: Advances in underwater inspection and maintenance. Aberdeen: Springer-Verlag GmbH.

Wawrzyniak N., Zaniewicz G., 2011: Wizualizacja ścian nabrzeży za pomocą obrazów sonaru skanującego na potrzeby geoinformatycznego systemu ochrony portu. Roczniki Geomatyki t.9, z.2(46), PTIP, Warszawa.

Xie S., Chen J., Luo J., Xie P., Tang W., 2012: Detection and Tracking of Underwater Object Based on Forward-Scan Sonar, 37506.