Underwater Mapping

We attended to the world’s premier event for marine science and ocean technology

Oceanology International is the world leading marine science and ocean technology exhibition. Every two years, Oceanology International brings together the latest technologies and worldwide leaders from a variety of industries including oil & gas, engineering, marine science and maritime security.   Girona Underwater Vision and Robotics team was present in the exhibition to showcase the two AUV prototypes developed by the research group (Girona 500 and Sparus II AUV) as well as the different image and sonar data processing capabilities.   The stand of the group received many interested visitors throughout the exhibition days thus enhancing the visibility of the group at a global scale and establishing new networking opportunities and  new relationships for potential future businesses.  

We attend the second Annual Catalan Meeting on Computer Vision (ACMCV)

The second Annual Catalan Meeting on Computer Vision (ACMCV) was held on the 18th of September in Universitat Pompeu Fabra of Barcelona. ACMCV was created with the aim of strengthen the network between computer vision researchers in Catalunya. As it is well known, Catalunya has a large community working in the computer vision field, both in the industry and the academia. Thus, this meeting is intended to get to know the people working in the field as well as their recent developments.   The event contained several activities, including a poster session, where the four members of Vicorob (Nuno Gracias, Josep Bosch, Konstantin Korotkov and Ricard Campos) presented their latest works. The meeting also held the Master thesis’ presentations of the students of the Master in Computer Vision of the Universitat Autonoma de Barcelona, as well as several interesting talks on topics such as the application of machine learning techniques…

Tali Hurtós defends her PhD thesis: “Forward-Looking Sonar Mosaicing for Underwater Environments”

Vehicle operations in underwater environments are frequently compromised by poor visibility conditions. The perception range of optical devices is heavily constrained in turbid waters, thus often complicating navigation and mapping tasks in environments such as harbors, bays, or rivers. A new generation of high-frequency forward-looking sonars that provide acoustic imagery at near-video frame rates have recently emerged as a promising alternative for working under these challenging conditions. In this thesis, we propose an end-to-end mosaicing framework tailored to the characteristics of forward-looking sonar imagery in order to build consistent overviews of planar underwater areas regardless of water visibility. Our solution targets versatility: it enables the generation of acoustic mosaics that involve roto-translational motions and comprise different vehicle tracklines; it is suitable for a wide range of scenarios, from feature-rich areas to environments with scarcity of features; it can be applicable on data collected with minimally instrumented vehicles; and it allows...
rafael Garcia

Rafael Garcia delivers a plenary talk at INT-NAM 2014

Rafael Garcia has delivered today the plenary lecture “3-D Mapping for the Characterization of the Seafloor using an Underwater Robot” as “keynote speaker”on the 2nd International Symposium on Maritime and Naval Architecture which takes place on October 23-24 in Istanbul, Turkey. In his keynote lecture, Dr. Garcia presented the latest advances and conclusions of the MICINN MUMAP project regarding the development of systems for seafloor monitoring.

arnau-udg-pandora

Open PhD position in Omnidirectional Computer Vision for Underwater Robots

The Computer Vision and Robotics Group of the University of Girona (UdG), Spain, has opened a full-time PhD position in the area of Omnidirectional Computer Vision for Underwater Robots. The research will be conducted in the Computer Vision and Robotics institute of the UdG (http://vicorob.udg.edu/) under the supervision of Dr. Rafael Garcia and Dr. Nuno Gracias.   The research will focus on using underwater cameras with very wide field of view (providing hemispherical coverage) to assist the operation of underwater robots. These cameras open up new unexplored capabilities of immersive telepresence, efficient optical mapping in cluttered environments and vehicle safety. The overall goal is to develop new capabilities to assist scientists operating a lightweight hybrid remotely operated vehicle in exploratory missions of mapping and surveying.   The group will offer a 4-year contract (3 years as PhD candidate + 1 year of postdoc).     Applicant profile   The applicants…

MORPH TRIALS

MORPH FP7 PROJECT TRIALS TAKE PLACE IN SANT FELIU DE GUÍXOLS (GIRONA)

From Wednesday 9th to Friday 18th of July, several MORPH European project experiments were held in Girona. Their goal was to prepare for the upcoming trials in the Azores islands next September. The Variable Geometry Range Only Navigation (VG-ROF) algorithm was tested with the IST Medusas, the UdG SparusII and Girona500, and the ATLAS SEACAT. The first three days were devoted to software integration, at CIRS facilities, to have everything ready for the trials carried out in Sant Feliu’s harbor. During the second week multiple missions with different vehicle configurations were tested. The main purpose was to grab images and multibeam data from the seafloor while maintaining the vehicle’s formation. Some tests regarding a new configuration, named USBL formation, were also simulated and tested in water.   The MORPH FP7 project is funded by the European Comission on ICT Robotics Programme.   Picture: DSOR Lab – IST – Portugal

EMRA2014

Presenting PANDORA FP7 and MORPH FP7 projects at EMRA 2014

Our researchers Pere Ridao and Marc Carreras are currently in Rome in the Workshop on EU Funded Marine Robotics and Applications (EMRA 2014). Researchers and users of marine robotic technology are invited to EMRA2014. The workshop shall summarize current EU FP7 marine robotics research, and provide a platform for marine stakeholders to share their current technological challenges. For researchers, EMRA2014 will offer dissemination opportunities for existing work, and highlight new application areas for consideration in future work. For marine research stakeholders, EMRA2014 will offer novel approaches to solve marine challenges, and a platform for directing future research threads. MORPH project will be presented today at 14h by Antonio Pascoal from IST-ISR and PANDORA project will be presented tomorrow in the afternoon by Marc Carreras. Follow ERMA2014 on streaming

3D mapping

Ricard Campos defends his PhD thesis “Surface Reconstruction Methods for Seafloor Modelling”

Underwater maps are an important source of information for the scientific community, since mapping the seafloor is the starting point for underwater exploration. The advance of range scanning methodologies, both using acoustic and optical techniques, enables the mapping of the seabed to attain increasingly larger resolutions. However, all these techniques sample the surface to reconstruct in the form of a point cloud. For the case of areas containing non-trivial 3D relief, achieving a continuous representation from this discrete sampling is a complex task. Surface reconstruction methods try to tackle this problem by recovering a continuous surface representing the object in the form of a mesh of triangles, easing visualization and further processing. This thesis proposes four different strategies to tackle the problem of surface reconstruction from point sets. We start by reviewing the state of the art on surface reconstruction methods. From this survey, we extract some conclusions regarding the…

ASM

Automated underwater object classification using optical imagery

PhD Thesis “Automated underwater object classification using optical imagery” By Shihavuddin, A.S.M Supervised by Dr. Nuno Grácias, Dr. Rafael García   Abstract This thesis addresses the problem of automated underwater optical image characterization. Remote underwater optical sensing allows the collection and storage of vast amounts of data for which manual classification may take months. Supervised automated classification of such datasets can save time and resources and can also enable extraction of valuable information related to marine and geological research.