Autonomous Optical Survey Based on Unsupervised Segmentation of Acoustic Backscatter

The application of acoustics to study the seabed have for decades provided industry and science with valuable information, and is still excels in terms of spatial coverage and detail. An acoustic response from the seabed not only contains information about the range, through the two way travel time, but also the acoustic reflectivity of the substrate from the strength of the backscatter response. As the signal strength differs between substrate types, this information can be used to detect and classify different seabed types. However, there are ambiguities in the acoustic signatures and the reliance on ground truth samples, for succeeding in this identification, is a limiting factor. In this paper we present a way to mitigate this problem using Hidden Markov Random Fields (HMRF) to perform unsupervised segmentation of the backscatter response for the purpose of determining different seabed types. The outcome of this analysis is directly used to plan and conduct an autonomous near-seabed camera survey to verify the classification results, whilst complementing the acoustical data-set. The method is tested in a full-scale experiment and performed in-situ onboard a Kongsberg Hugin 1000 autonomous underwater vehicle (AUV).

[1]  Carl E. Rasmussen,et al.  Factorial Hidden Markov Models , 1997 .

[2]  Stan Z. Li.  Markov Random Field Modeling in Image Analysis , 2009, Advances in Pattern Recognition.

[3]  Y. Marcon,et al.  The use of multibeam backscatter and bathymetry as a means of identifying faunal assemblages in a deep-sea cold seep , 2016 .

[4]  John R. Hershey,et al.  Variational Bhattacharyya divergence for hidden Markov models , 2008, 2008 IEEE International Conference on Acoustics, Speech and Signal Processing.

[5]  Stephen M. Smith,et al.  Segmentation of brain MR images through a hidden Markov random field model and the expectation-maximization algorithm , 2001, IEEE Transactions on Medical Imaging.

[6]  N. Gracias,et al.  Hydrothermal activity along the slow-spreading Lucky Strike ridge segment (Mid-Atlantic Ridge): Distribution, heatflux, and geological controls , 2015 .

[7]  Christian Hennig,et al.  Methods for merging Gaussian mixture components , 2010, Adv. Data Anal. Classif..

[8]  Hagai Attias,et al.  A Variational Bayesian Framework for Graphical Models , 1999 .

[9]  Pushmeet Kohli,et al.  Markov Random Fields for Vision and Image Processing , 2011 .

[10]  Deborah K. Smith,et al.  Fault rotation and core complex formation: Significant processes in seafloor formation at slow‐spreading mid‐ocean ridges (Mid‐Atlantic Ridge, 13°–15°N) , 2008 .

[11]  S. Kasten,et al.  Current status of manganese nodule exploration in the German license area , 2011 .

[12]  Peter Clifford,et al.  Markov Random Fields in Statistics , 2012 .

[13]  Ricardo Martins,et al.  The LSTS toolchain for networked vehicle systems , 2013, 2013 MTS/IEEE OCEANS - Bergen.

[14]  Martin Syre Wiig,et al.  NTNU-FFI Cruise 2017 - Hugin Autonomy Integration (DUNE, T-REX) , 2017 .

[15]  Robert Grover Brown,et al.  Introduction to random signals and applied Kalman filtering : with MATLAB exercises and solutions , 1996 .

[16]  D. Orange,et al.  Improved detection and mapping of deepwater hydrocarbon seeps: optimizing multibeam echosounder seafloor backscatter acquisition and processing techniques , 2018, Marine Geophysical Research.

[17]  Jean-Marie Augustin,et al.  Analysis of seafloor backscatter strength dependence on the survey azimuth using multibeam echosounder data , 2017, Marine Geophysical Research.

[18]  J. Clarke.  Multispectral Acoustic Backscatter from Multibeam , Improved Classification Potential , 2015 .

[19]  V. Guida,et al.  Sedimentary facies, geomorphic features and habitat distribution at the Hudson Canyon head from AUV multibeam data , 2015 .

[20]  Robert A. Altenburg,et al.  Acoustic backscattering at low grazing angles from the ocean bottom. Part II. Statistical characteristics of bottom backscatter at a shallow water site , 1984 .

[21]  Alexandre C. G. Schimel,et al.  Backscatter measurements by seafloor‐mapping sonars: guidelines and recommendations , 2015 .

[22]  Yee Whye Teh,et al.  Bayesian Nonparametric Models , 2010, Encyclopedia of Machine Learning.