OCEANS'13 San Diego

Christophe Sintes, OES Student Activities Coordinator, Photos by Stan Chamberlain

The 33rd Student Poster Program of the OCEANS Conferences was held at the OCEANS’13 MTS/IEEE San Diego conference at the Town and Country Resort, San Diego, USA, September 23–26, 2013. Once again outstanding posters desrcibed the work that the students were presenting. The program was organized and directed by Kevin Delaney as local coordinator and by the student activities coordinators of both organizing societies, Jill Zande from MTS and Christophe Sintes from IEEE OES. 16 student posters were accepted from the 100 abstracts received. Once again the program was supported by funding from the US Navy Office of Naval Research, which enabled the students to attend the conference.
     The roster of students and their schools are:

  • Murat Aykin, University of Miami
  • Daniel Coles, University of Southampton
  • Bruno Ferreira, Instituto de Engenharia de Sistemas e Computadores do Porto (INESC TEC)
  • Donya Frank, University of New Hampshire
  • Akihisa Fukami, University of Tokyo
  • Natalia Hurtos, University of Girona
  • Wen Liu, Hangshou Dianzi University
  • Masaru Nagaso, Institute of Industrial Science, University of Tokyo
  • Samir Ouelha, DCNS
  • Eric Piper, Florida State University
  • Edward Richards, University of Rhode Island
  • Luke Rumbaugh, Clarkson University
  • Ken Sooknanan, Trinity College
  • Tomoko Takahashi, University of Tokyo
  • Christopher Tomaszewski, Robotics Institute, Carnegie Mellon University
  • Sean Walstead, University of California, San Diego

     The posters were judged by a team organized by MTS and IEEE OES. The student award winners were announced at the Thursday exhibitor’s luncheon. Dr. Sintes opened the awards ceremony and introduced Jill Zande and Kevin Delaney, who presented each student with a Certificate of Participation in the OCEANS’13 MTS/IEEE San Diego. The students were than all introduced as a group and received a round of applause from the conference attendees. The students were announced as members of the “OCEANS Student Poster Alumni Association”. This year Norman Miller could not attend the conference; to commemorate his founding of the poster competition and the 600th student participating in the Student poster competition, MTS and IEEE OES have decided to name the first prize after him.
     René Garello and Drew Michel, presidents of IEEE OES and MTS, announced the winning student posters and presented the awards to the students.

Society Presidents and Student Poster Competition winners
(l to r): Drew Michel, Sean Walstead, Donya Frank,
Natalia Hurtos, Luke Rumbaugh, René Garello.
  • The honorable mention prize was given to Donya Frank.
  • The third prize was given to Sean Walstead who received an award of $1,000 for his research.
  • The second prize was presented to Luke Rumbaugh with an award of $2,000 for his work on laser for underwater remote sensing.
  • The first prize, the Norman Miller award, with $3,000 was presented to Natalia Hurtos for her research in sonar processing.

     The winning students all received a round of applause for their accomplishments and participation in the Student Poster Program of SanDiego!
     The abstracts for each of the presented posters are reprinted below. The full poster paper of the first place winner is included following this article.


Winning Poster Paper

Murat Aykin, University of Miami, Forward-Look 2-D Sonar Image Formation and 3-D Reconstruction
Abstract—Sonar imagery deals with transmitting acoustic signals and measuring the reflected sound from the scene surfaces. The recorded signal encodes information about the shape and material properties of these surfaces. The relationship of the scene geometry to the image irradiance can be established by modeling the physics of the sonar image formation process. The inherent ambiguities in the interpretation of the 3-D world based on visual cues in a 2-D forward-scan sonar image arise as a result of both operating as a ranging device and loss of elevation angle information due to the projection geometry. One aspect of this work is modeling the image formed by a new class of high-resolution 2-D forward-looking sonar systems which supports our complementary aim of recovering the unknown zenith angles from image brightness and thus reconstruct 3-D objects. Our method applies to a single forward-scan sonar image, assuming that the scene objects have smooth surfaces that vary monotonically in terms of distance from the sonar, and cast visible shadows on a flat background. We present the results of experiments with real data to demonstrate the performance of our 3-D reconstruction technique.


Daniel Coles, University of Southampton, Resource Assessment of Large Marine Current Turbine Arrays
Abstract—At present, simple analytical and numerical models exist that give approximations for the energy that can be extracted by marine current turbine arrays (MCTAs) from an available resource. This paper first presents an analysis of different methods for conducting a resource assessment to outline the considerations that must be made whilst also ­identifying areas where more detail is required. Resource assessment using 2-D depth averaged numerical modeling software Telemac-2D is presented to give analysis of the behaviour of the flow in the presence of MCTAs and the effects of array density and seabed roughness on wake length and power generation.



Bruno Ferreira, Instituto de Engenharia de Sistemas e Computadores do Porto (INESC TEC), Localization of a sound source: optimal positioning of sensors carried on autonomous surface vehicles
This paper addresses the problem of optimal, three-dimensional, localization of an acoustic pinger. Making use of acoustic receivers, the time-of-arrivals are computed and fused with their positions to estimate the target position. Motivated by practical applications where autonomous surface vehicles are employed to carry sensing equipment, these receivers are constrained to lie in a plane. The optimal configuration of sensors is derived by maximizing the determinant of the Fisher information matrix. A method to track and estimate the position of the target is proposed and implemented in a formation of four vehicles. Experimental results show very motivating results with successful estimates of the target position.


Donya Frank, University of New Hampshire, Direct Measurements of Sediment Response to Waves with “Smart Sediment Grains”
Measurements of sediment motion have been primarily limited to indirect observations with acoustic and optical instruments. A micro-electro-mechanical systems device that can measure and record Lagrangian observations of coastal sediments at incipient motion has been developed. These sensors move freely, measure acceleration in the six degrees of freedom and their mobility characteristics are similar to coarse gravel. Experiments conducted in a small oscillating flow tunnel and a large wave flume verified that the sensors detect incipient motion under various hydrodynamic conditions. Analysis of complementary fluid velocity measurements suggests the influence of pressure gradient induced sediment motion.



Akihisa Fukami, University of Tokyo, Sub-Bottom Synthetic Aperture Sonar in Robust Circumstances
In Kanda Bay of Fukuoka, Japan, 2,996 pieces of chemical ammunitions were detected in the exploration until 2012. Therefore, we have developed sub-bottom synthetic aperture sonar for high resolution at low frequency to explore the targets of the deposited layer. Sonar can be mounted on a small ship under the robust circumstances of environment. We have considered next five issues; refraction at the deposited layer, the frequent velocity change due to tide and wave for small ships, also frequent movement of sensors, weak signals in the deposited layer, and flexibility in offline processing, and have developed the solutions. We have carried out the detection test of targets buried in the seabed sediments at Kanda Bay in July and September 2012, and we report the processing method and the result in this paper.


Natalia Hurtos, University of Girona, A Novel Blending Technique for Two-Dimensional Forward-Looking Sonar Mosaicing
High-resolution forward-looking sonars are becoming a tool of choice for exploring underwater environments under low visibility conditions. Their imagery can be mosaicked to obtain a global overview of submerged areas of interest and the spatial arrangement of different target features. However, in order to achieve an informative and smooth image composition, the individual sonar frames must be fused. Unlike the blending in optical mosaics, this implies dealing with a high number of overlapping images as well as with sonar specific artifacts arising from its image formation geometry. This work presents a novel blending pipeline designed to cope with these artifacts involving strategies to diminish the impact of all the photometric irregularities that might be present when mosaicing forward looking sonar imagery. Results of blended mosaics, ­including data gathered with different sonar models and presenting several artifacts, are presented here to show the applicability of the method.




Wen Liu, Hangshou Dianzi University, Brushless DC motor control system based on submarine hybrid transmission technology
The energy and data hybrid transmission technology is one of the most important technologies in the field of submarine investigation and exploitation. It is urgent to study high-power and high-speed hybrid transmission technology. All kinds of submarine equipment, using electric motors to drive the actuating mechanisms, require the kind of technology to solve the energy and data hybrid transmission problem. In order to solve this problem, our research team designed and verified one kind of submarine motor control system, which is based on the Hybrid transmission technology and BLDC (Brushless Direct Current) motor control technology. A single coaxial cable, with a length of 10 km, is used for the high-voltage DC power source transmission. The high-speed data is coupled on the same cable by capacitance. The system can supply 3 KW power to submarine equipment. It has achieved good results that the platform on the deck can monitor and control underwater motor system in real time. The experiments show that the system is feasible and practical. It will be used by all kinds of underwater equipments widely.


Masaru Nagaso, Institute of Industrial Science, University of Tokyo, Development of the three-dimensional visualization method for the inner structure of small size fish using 25 MHz acoustic profile measurement
A visualization method using ultrasonic acoustic profile measurement was introduced and inner structures of small size fishes were examined. Detection of external and internal body surface were executed. Two kinds of fishes are chosen in this experiment and measured using a 25 MHz medical probe. Then signal processing and visualization of three dimensional structure including inner part of the body, abstraction of the skin and organ boundary surface and ­mapping of reflection signal strength on their surfaces with corrections for surface gradient acoustical loss and scattering attenuation by referring the reflection strength change rate data, relating on the gradient and distance from the probe to target surface. Data for correction was taken from cylindrical silicon model are carried out. Three-dimensional acoustic images are obtained and the result shows that the detection of organs is possible even if these are small size organs of small size fishes. Then the reflection signals from the boundary surface of organs could also be seen clearly. Surface abstraction and reflection strength mapping shows points which has strong hardness and has great change of acoustic impedance. Application for fish species classification by reflection signal amplitude on body surface were tried and its capability was confirmed.


Samir Ouelha, DCNS, Extension of maximal marginal diversity based feature selection applied to underwater acoustic data
This paper addresses the feature selection problem encountered in underwater acoustic data mining. Feature selection is a preamble of any data mining algorithm, allowing a priori dimension reduction and better interpretation of data. Here, we propose a new feature selection technique, based on the maximum marginal diversity principle. Our approach is applied on various real dataset, including underwater acoustic data.





Eric Piper, Florida State University, Oil Droplet Fate in the Gulf of Mexico
The fate of the oil from the Deepwater Horizon blowout can be best understood from the perspective of small oil droplets. These were essentially created deep in the well when the methane gas came out of solution due to reduced pressures. This caused a very turbulent flow with resulting high shearing forces creating a mixture of methane gas and generally microscopic sized oil droplets. Although initially buoyed upward due to the methane gas bubbles entrained in the flow, these oil droplets were eventually transported according to Stokes’ Law and ocean currents. Along the way, the oil droplets tended to get filtered out of the water column through a variety of dynamic physical and biological processes.




Edward Richards, University of Rhode Island, Range Limitations on Acoustic Adjoint Inversions
Measurement of range dependent ocean sound speed is an important to many ocean acoustic problems, including source localization and tomography. The recent prevalence of low cost robotic platforms such as oceanographic gliders has increased the availability of long-term measurements of the ocean environment, but in many cases their usefulness is limited by the spatial-temporal resolution of their measurements. Combining them with additional acoustic inversion sampling of the same sound speed field may increase these measurements’ resolution. This paper investigates a method of combining the adjoint method, a local acoustic inversion, with glider measurements to create reliable sound speed measurements at high sampling rates. The performance of this method is found to be highly sensitive to the total range study area with a simulation study comparing root mean squared error over 30 simulated inversions. This limitation may be an effect of the limited ensonification of the water column at longer ranges, which is shown using a ray trace model.


Luke Rumbaugh, Clarkson University, A 532 nm Chaotic Lidar Transmitter for High Resolution Underwater Ranging and Imaging
A chaotic lidar transmitter based on an ultralong cavity fiber laser is presented for underwater ranging and imaging applications. Experimental results show 150 mW continuous output power at 532 nm, with wideband (>3 GHz) chaotic intensity modulation. An incoherent digital receiver is used to show a ranging accuracy of 1 cm in a water tank with a measurement-limited range resolution of –4 cm. There is no range ambiguity because of the non-repeating nature of the chaotic signal.




Ken Sooknanan, Trinity College, Mosaics For Burrow Detection in Underwater Surveillance Video
Harvesting the commercially significant lobster, Nephrops norvegicus, is a multimillion dollar industry in Europe. Stock assessment is essential for maintaining this activity but it is conducted by manually inspecting hours of underwater surveillance videos. To improve this tedious process, we propose the use of mosaics for the automated detection of burrows on the seabed. We present novel approaches for handling the difficult lighting conditions that cause poor video quality in this kind of video material. Mosaics are built using 1–10 minutes of footage and candidate burrows are selected using image segmentation based on local image contrast. A K-Nearest Neighbour classifier is then used to select burrows from these candidate regions. Our final decision accuracy at 93.6% recall and 86.6% precision shows a corresponding 18% and 14.2% improvement compared with previous work [1].



Tomoko Takahashi, University of Tokyo, Investigation of Long-Pulse Laser-Induced Breakdown Spectroscopy for Analysis of the Composition of Rock and Sediment Samples Submerged in Seawater
The aim of this paper is to investigate the application of Laser-Induced Breakdown Spectroscopy (LIBS) to qualitatively analyze the composition of various seafloor rock samples in seawater. Well-resolved emission spectra of submerged samples were achieved by using a long laser pulse of duration 250 ns. It was found that the main elements in each rock sample could be successfully identified both in pure water and in seawater. When comparing the measurements made in pure water and in seawater, some effects of the seawater could be seen, however these effects are negligible and do not have any significant detrimental effect on the analytical value of the signal obtained. The results suggest that long-pulse LIBS may be applicable for in situ, multi-element chemical analysis of sediments and rocks in the marine environment.


Christopher Tomaszewski, Robotics Institute, Carnegie Mellon University, Planning Efficient Paths through Dynamic Flow Fields in Real World Domains
This research addresses the problem of planning efficient paths for agents through flow fields in small real-world domains where vehicle dynamics and environmental uncertainty can significantly affect the optimality of a path. In particular, we consider the task of planning routes for small autonomous airboats deployed in various river domains so as to best take advantage of water currents to save energy and time. Existing planning techniques for flow fields were implemented on our airboat platform and evaluated on these domains with current models developed from the data gathered using a Nortek AD2CP-Glider acoustic doppler current profiler. The real-world performance of these algorithms were compared to theoretical estimates and several modifications are suggested to improve their performance in specific domains.



Sean Walstead, University of California, San Diego, Reconstructing surface wave profiles from forward scattered pulses
Surface wave shape is determined by analyzing underwater reflected acoustic signals collected at multiple receivers in a wave tank. The acoustic signals are short 3-cycle pulses of nominal frequency 300 kHz. With the use of a forward scattering model based on the Kirchhoff approximation, an inverse processing algorithm determines surface wave shape by matching model and data acoustic waveforms. Reconstructed surfaces are accurate to a resolution of about a quarterwavelength of the acoustic pulse only within Fresnel zones associated with each source and receiver pair. Within Fresnel zone regions, the acoustically derived surface is known with greater certainty than the surface profile as measured with a camera. Multiple receivers extend the region of surface reconstructed acoustically. The type of wave features to be investigated by this technique is scalable with acoustic transmission frequency. This versatility in implementation leads to the prospect of ocean application across a wide range of length scales.