Pub Date : 2019-06-17DOI: 10.1109/OCEANSE.2019.8867293
G. Matte, F. Chaillan, A. Heinzle
The goal of this study is to take advantage of artificial intelligence (AI) algorithms and distributed computing ecosystem to enhance the 3D-multibeam echo sounder data processing functionality. We consider first the post processing case, where a complete dataset has been recorded during sea trials. Hence, our suggested framework designed for massive real world data processing allows employing false alarm reduction and underwater object recognition techniques, which can be easily used as decision making for underwater autonomous vehicle safe navigation.
{"title":"3D Multibeam Echo Sounder Data Processing Using Distributed Computing. Application To False Alarm Reduction And Unsupervised Underwater Object Recognition For Safe Navigation","authors":"G. Matte, F. Chaillan, A. Heinzle","doi":"10.1109/OCEANSE.2019.8867293","DOIUrl":"https://doi.org/10.1109/OCEANSE.2019.8867293","url":null,"abstract":"The goal of this study is to take advantage of artificial intelligence (AI) algorithms and distributed computing ecosystem to enhance the 3D-multibeam echo sounder data processing functionality. We consider first the post processing case, where a complete dataset has been recorded during sea trials. Hence, our suggested framework designed for massive real world data processing allows employing false alarm reduction and underwater object recognition techniques, which can be easily used as decision making for underwater autonomous vehicle safe navigation.","PeriodicalId":375793,"journal":{"name":"OCEANS 2019 - Marseille","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122476155","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-06-17DOI: 10.1109/OCEANSE.2019.8867059
A. Morozov, D. Webb
There is a growing interest for a very low frequency sound source in the frequency range below 20 Hz for such applications as Arctic under-ice acoustic, far-range navigation, communications and Thermometry, sub-bottom seismic profiling, et cetera. The ultra-low frequency sound propagates without attenuation and loss of coherency at a very far distance covering the water column from the surface to the ocean floor. Another aspect of the same problem, which has been in an increasing focus of oil and gas producers, is the reducing the impact of noise from traditional air-guns on marine mammals. A coherent sound source can be a quieter and more benign to marine mammals. The major oil companies, Shell, Exxon and Total, are sponsoring are sponsoring the Marine Vibrator Joint Industry Project MVJIP. Marine Vibrators are a coherent type of seismic source, which less harmful for marine inhabitants and gives a clearer, more precise and higher resolution imaging of the bottom formations, structures, and deposits. Teledyne Webb Research is one of the participants in the MVJIP. Teledyne Webb Research has many years of experience in a deep water sound source development showing that to build a sound source with a frequency below 20 Hz is a hard task due to a very large emitted volume velocity or product of aperture area to its linear displacement. For sound pressure level (SPL) larger than 200 dB re 1 uPa at 1 meter the volume displacement at 5 Hz cycle can be tens of liters. Systems with rigid or flexural vibrating diaphragm with a large aperture area are difficult to build, and usually not efficient or have a very narrow bandwidth. Highly efficient frequency sweeping sound sources on the base of tunable organ pipes show very good performance for 150 - 2000 Hz frequency bandwidth. This technology can potentially reach a frequency bandwidth 70-100 Hz, while keeping a high efficiency performance. However, a further decrease of the frequency will be hard to achieve because of the organ pipe growing dimension. The expected complication from such giant design demands us to look for other more simple approaches for underwater sound emitting. As one of the participants in the Marine Vibrator JIP, Teledyne Webb Research is developing a coherent seismic marine sound source technology based on the application of an underwater, gas filled bubble resonator as a very low frequency seismic source. This innovative system is a promising candidate for a high power, highly efficient, and coherent seismic source. The gas-filled bubble offers the large radiating area and was shown to be a good impedance transformer with very high radiation efficiency. The bubble sound source has a simple design using a standard commercial off-the-shelf driver. The elastic membrane supports high volume displacement with a large radiation aperture and prevents cavitation damage. Large volume displacement and velocity support the large radiation power. The sound sources have very small coupling effe
{"title":"Underwater Infra-Sound Resonator for Long Range Acoustics and Seismic Survey","authors":"A. Morozov, D. Webb","doi":"10.1109/OCEANSE.2019.8867059","DOIUrl":"https://doi.org/10.1109/OCEANSE.2019.8867059","url":null,"abstract":"There is a growing interest for a very low frequency sound source in the frequency range below 20 Hz for such applications as Arctic under-ice acoustic, far-range navigation, communications and Thermometry, sub-bottom seismic profiling, et cetera. The ultra-low frequency sound propagates without attenuation and loss of coherency at a very far distance covering the water column from the surface to the ocean floor. Another aspect of the same problem, which has been in an increasing focus of oil and gas producers, is the reducing the impact of noise from traditional air-guns on marine mammals. A coherent sound source can be a quieter and more benign to marine mammals. The major oil companies, Shell, Exxon and Total, are sponsoring are sponsoring the Marine Vibrator Joint Industry Project MVJIP. Marine Vibrators are a coherent type of seismic source, which less harmful for marine inhabitants and gives a clearer, more precise and higher resolution imaging of the bottom formations, structures, and deposits. Teledyne Webb Research is one of the participants in the MVJIP. Teledyne Webb Research has many years of experience in a deep water sound source development showing that to build a sound source with a frequency below 20 Hz is a hard task due to a very large emitted volume velocity or product of aperture area to its linear displacement. For sound pressure level (SPL) larger than 200 dB re 1 uPa at 1 meter the volume displacement at 5 Hz cycle can be tens of liters. Systems with rigid or flexural vibrating diaphragm with a large aperture area are difficult to build, and usually not efficient or have a very narrow bandwidth. Highly efficient frequency sweeping sound sources on the base of tunable organ pipes show very good performance for 150 - 2000 Hz frequency bandwidth. This technology can potentially reach a frequency bandwidth 70-100 Hz, while keeping a high efficiency performance. However, a further decrease of the frequency will be hard to achieve because of the organ pipe growing dimension. The expected complication from such giant design demands us to look for other more simple approaches for underwater sound emitting. As one of the participants in the Marine Vibrator JIP, Teledyne Webb Research is developing a coherent seismic marine sound source technology based on the application of an underwater, gas filled bubble resonator as a very low frequency seismic source. This innovative system is a promising candidate for a high power, highly efficient, and coherent seismic source. The gas-filled bubble offers the large radiating area and was shown to be a good impedance transformer with very high radiation efficiency. The bubble sound source has a simple design using a standard commercial off-the-shelf driver. The elastic membrane supports high volume displacement with a large radiation aperture and prevents cavitation damage. Large volume displacement and velocity support the large radiation power. The sound sources have very small coupling effe","PeriodicalId":375793,"journal":{"name":"OCEANS 2019 - Marseille","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115423827","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-06-17DOI: 10.1109/OCEANSE.2019.8867478
I. Stanković, C. Ioana, M. Daković
An analysis of different sequences for the reconstruction and targeting of underwater sonar images is presented. The sonar images are assumed to be sparse, and their reconstruction is possible by using the compressive sensing theory. The goal is to localize and reconstruct targets by using an iterative version of the orthogonal matching pursuit (OMP) method. The sequences which are used as the transmitted signal waveforms are formed with: the Alltop sequence, the M sequence, a random Gaussian sequence, a binary random sequence, the Zadoff-Chu sequence, and the Bjorck sequence. The comparison of the reconstruction results is done for various numbers of samples in the sequences and sparsity levels. An analysis of the performance for each of the sequences in various noise levels is done as well. Percentage of successfully detected targets is used as a performance measure.
{"title":"Sequence Comparison in Reconstruction and Targeting in Underwater Sonar Imaging","authors":"I. Stanković, C. Ioana, M. Daković","doi":"10.1109/OCEANSE.2019.8867478","DOIUrl":"https://doi.org/10.1109/OCEANSE.2019.8867478","url":null,"abstract":"An analysis of different sequences for the reconstruction and targeting of underwater sonar images is presented. The sonar images are assumed to be sparse, and their reconstruction is possible by using the compressive sensing theory. The goal is to localize and reconstruct targets by using an iterative version of the orthogonal matching pursuit (OMP) method. The sequences which are used as the transmitted signal waveforms are formed with: the Alltop sequence, the M sequence, a random Gaussian sequence, a binary random sequence, the Zadoff-Chu sequence, and the Bjorck sequence. The comparison of the reconstruction results is done for various numbers of samples in the sequences and sparsity levels. An analysis of the performance for each of the sequences in various noise levels is done as well. Percentage of successfully detected targets is used as a performance measure.","PeriodicalId":375793,"journal":{"name":"OCEANS 2019 - Marseille","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129476308","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-06-17DOI: 10.1109/OCEANSE.2019.8867290
G. Acosta, B. menna, Ignacio Carlucho, M. D. Paula, S. Villar, H. Curti, A. Rozenfeld, Roberto J. de la Vega, Agustin Isasmendi, R. C. Leegstra, Luis M. Arrien
Autonomous Underwater Vehicles (AUVs) are suitable platforms for a wide type of applications in the oceanic environment. These applications are developed in various fields such as scientific surveying, off-shore industry and defense. The employment of AUVs requires less human support and reduces operation costs. Due to the changing marine environment these vehicles must deal with uncertain and hostile conditions to perform its tasks. In the marine robotics matter, the INTELYMEC group has developed in 2012 an AUV prototype called Ictiobot, a low cost experimental platform for multipurpose missions. In this paper an upgrade of the original prototype is presented, the Ictiobot-40, conceived to perform acoustic imaging surveying missions of up to two hours and maximum depths of 40 meters. The new software and hardware architectures and mechanical structure improvements, are detailed. In addition to these technical details, initial experimental results of the AUV performance in quiet waters will be discussed. Also, the new approaches for systems under development are presented.
{"title":"Ictiobot-40 a low cost AUV platform for acoustic imaging surveying","authors":"G. Acosta, B. menna, Ignacio Carlucho, M. D. Paula, S. Villar, H. Curti, A. Rozenfeld, Roberto J. de la Vega, Agustin Isasmendi, R. C. Leegstra, Luis M. Arrien","doi":"10.1109/OCEANSE.2019.8867290","DOIUrl":"https://doi.org/10.1109/OCEANSE.2019.8867290","url":null,"abstract":"Autonomous Underwater Vehicles (AUVs) are suitable platforms for a wide type of applications in the oceanic environment. These applications are developed in various fields such as scientific surveying, off-shore industry and defense. The employment of AUVs requires less human support and reduces operation costs. Due to the changing marine environment these vehicles must deal with uncertain and hostile conditions to perform its tasks. In the marine robotics matter, the INTELYMEC group has developed in 2012 an AUV prototype called Ictiobot, a low cost experimental platform for multipurpose missions. In this paper an upgrade of the original prototype is presented, the Ictiobot-40, conceived to perform acoustic imaging surveying missions of up to two hours and maximum depths of 40 meters. The new software and hardware architectures and mechanical structure improvements, are detailed. In addition to these technical details, initial experimental results of the AUV performance in quiet waters will be discussed. Also, the new approaches for systems under development are presented.","PeriodicalId":375793,"journal":{"name":"OCEANS 2019 - Marseille","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128029620","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-06-17DOI: 10.1109/OCEANSE.2019.8867572
V. Rérolle, D. Angelescu, A. Hausot, P. Ea, N. Lefèvre, C. Provost, M. Labaste
Ocean acidification is a direct consequence of the atmospheric CO2 increase and represents a threat for marine ecosystems, particularly in the Arctic. High-quality seawater pH measurements with good spatial and temporal coverage are required to apprehend the ocean acidification phenomena. We are working to develop a high-accuracy, high-resolution pH sensor that has the potential to allow global ocean acidification mapping through deployment on fleets of ARGO floats and other autonomous platforms already in existence. The instrument implements a novel hybrid approach, utilizing the two different and complementary measurement techniques (potentiometric and colorimetric) to generate temporally dense and highly accurate pH data. Here we present the concept and initial results obtained from a hybrid pH sensor. Results show that the potentiometric part of the sensor is capable to operate in real ocean pressure and temperature conditions, including near-freezing temperatures typical of Arctic environmental conditions. The colorimetric part provides a stable reference to perform periodic recalibrations and remove drift.
{"title":"Development of a novel hybrid pH sensor for deployment on autonomous profiling platforms","authors":"V. Rérolle, D. Angelescu, A. Hausot, P. Ea, N. Lefèvre, C. Provost, M. Labaste","doi":"10.1109/OCEANSE.2019.8867572","DOIUrl":"https://doi.org/10.1109/OCEANSE.2019.8867572","url":null,"abstract":"Ocean acidification is a direct consequence of the atmospheric CO2 increase and represents a threat for marine ecosystems, particularly in the Arctic. High-quality seawater pH measurements with good spatial and temporal coverage are required to apprehend the ocean acidification phenomena. We are working to develop a high-accuracy, high-resolution pH sensor that has the potential to allow global ocean acidification mapping through deployment on fleets of ARGO floats and other autonomous platforms already in existence. The instrument implements a novel hybrid approach, utilizing the two different and complementary measurement techniques (potentiometric and colorimetric) to generate temporally dense and highly accurate pH data. Here we present the concept and initial results obtained from a hybrid pH sensor. Results show that the potentiometric part of the sensor is capable to operate in real ocean pressure and temperature conditions, including near-freezing temperatures typical of Arctic environmental conditions. The colorimetric part provides a stable reference to perform periodic recalibrations and remove drift.","PeriodicalId":375793,"journal":{"name":"OCEANS 2019 - Marseille","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122409021","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-06-17DOI: 10.1109/OCEANSE.2019.8867434
P. Cieślak
The marine robotics community is lacking a high quality simulator for doing scientific research, especially when it comes to testing control and vision algorithms in realistic underwater intervention tasks. All of the solutions used today are either outdated or try to combine different software tools, which often results in bad performance, stability issues and lack of important features. This paper presents a new software tool, focused on, but not limited to, simulation of intervention autonomous underwater vehicles (I-AUV). It delivers advanced hydrodynamics based on actual geometry, simulation of underwater sensors and actuators, as well as realistic rendering of underwater environment and ocean surface. It consists of a library written in C++ and a Robot Operating System (ROS) package.
{"title":"Stonefish: An Advanced Open-Source Simulation Tool Designed for Marine Robotics, With a ROS Interface","authors":"P. Cieślak","doi":"10.1109/OCEANSE.2019.8867434","DOIUrl":"https://doi.org/10.1109/OCEANSE.2019.8867434","url":null,"abstract":"The marine robotics community is lacking a high quality simulator for doing scientific research, especially when it comes to testing control and vision algorithms in realistic underwater intervention tasks. All of the solutions used today are either outdated or try to combine different software tools, which often results in bad performance, stability issues and lack of important features. This paper presents a new software tool, focused on, but not limited to, simulation of intervention autonomous underwater vehicles (I-AUV). It delivers advanced hydrodynamics based on actual geometry, simulation of underwater sensors and actuators, as well as realistic rendering of underwater environment and ocean surface. It consists of a library written in C++ and a Robot Operating System (ROS) package.","PeriodicalId":375793,"journal":{"name":"OCEANS 2019 - Marseille","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127509859","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-06-17DOI: 10.1109/OCEANSE.2019.8867251
Marion Poupard, Paul Best, Jan Schlüter, Jean-Marc Prevot, H. Symonds, P. Spong, H. Glotin
One of the best ways of studying animals that produce signals in underwater environments is to use passive acoustic monitoring (PAM). Acoustic monitoring is used to study marine mammals in oceans, and gives us information for understanding cetacean life, such as their behaviour, movement or reproduction. Automated analysis for captured sound is almost essential because of the large quantity of data. A deep learning approach was chosen for this task, since it has proven great efficiency for answering such problematics. This study focused on the orcas (Orcinus orca) of northern Vancouver Island, Canada, in collaboration with the NGO Orcalab which developed a multi-hydrophone recording station around Hanson Island to study orcas. The acoustic station is composed of 5 hydrophones and extends over 50 km2 of ocean. Since 2016 we are continuously streaming the hydrophone signals to our laboratory at Toulon, France, yielding nearly 50 TB of synchronous multichannel recordings. The objective for this research is to do a preliminary analysis of the collected data and demonstrate influence of environmental factors (tidal, moon phase and daily period) on the orcas' acoustic activities.
{"title":"Deep Learning for Ethoacoustics of Oreas on three years pentaphonie continuous recording at Orealab revealing tide, moon and diel effects","authors":"Marion Poupard, Paul Best, Jan Schlüter, Jean-Marc Prevot, H. Symonds, P. Spong, H. Glotin","doi":"10.1109/OCEANSE.2019.8867251","DOIUrl":"https://doi.org/10.1109/OCEANSE.2019.8867251","url":null,"abstract":"One of the best ways of studying animals that produce signals in underwater environments is to use passive acoustic monitoring (PAM). Acoustic monitoring is used to study marine mammals in oceans, and gives us information for understanding cetacean life, such as their behaviour, movement or reproduction. Automated analysis for captured sound is almost essential because of the large quantity of data. A deep learning approach was chosen for this task, since it has proven great efficiency for answering such problematics. This study focused on the orcas (Orcinus orca) of northern Vancouver Island, Canada, in collaboration with the NGO Orcalab which developed a multi-hydrophone recording station around Hanson Island to study orcas. The acoustic station is composed of 5 hydrophones and extends over 50 km2 of ocean. Since 2016 we are continuously streaming the hydrophone signals to our laboratory at Toulon, France, yielding nearly 50 TB of synchronous multichannel recordings. The objective for this research is to do a preliminary analysis of the collected data and demonstrate influence of environmental factors (tidal, moon phase and daily period) on the orcas' acoustic activities.","PeriodicalId":375793,"journal":{"name":"OCEANS 2019 - Marseille","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130996649","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-06-17DOI: 10.1109/OCEANSE.2019.8867040
X. Bompais, S. Garziglia, J. Blandin, Y. Hello
Thanks to a fruitful collaboration between Ifremer (Institut français pour l’exploitation de la mer), Geoazur and CNRS (Centre national de la recherche scientifique), a subsea cabled observatory has been operating offshore Nice (France) since 2015. It is intended to collect measurements of several instruments (seismometer and piezometers) in an area recognized as a natural laboratory for the study of the underlying mechanisms of submarine slope failures. The observatory is made up of two subsea stations, connected to a shore station by an electro-optical cable. All the components of the infrastructure and instruments can be remotely accessed through the internet. The measurements of the instruments are automatically transmitted to data centers. They can be visualized and downloaded on web portals.
得益于Ifremer (Institut franais pour l’exploitation de la mer)、Geoazur和CNRS (Centre national de la recherche scientifique)之间富有成效的合作，海底电缆观测站自2015年起在法国尼斯(Nice)近海运行。它的目的是在一个被认为是研究海底斜坡破坏潜在机制的天然实验室的地区收集几种仪器(地震仪和气压计)的测量结果。该观测站由两个海底观测站组成，通过一条光电电缆与岸上观测站相连。基础设施和仪器的所有组成部分都可以通过互联网远程访问。仪器的测量结果自动传输到数据中心。它们可以在门户网站上可视化和下载。
{"title":"EMSO-Ligure Nice, a Coastal Cabled Observatory Dedicated to the Study of Slope Stability","authors":"X. Bompais, S. Garziglia, J. Blandin, Y. Hello","doi":"10.1109/OCEANSE.2019.8867040","DOIUrl":"https://doi.org/10.1109/OCEANSE.2019.8867040","url":null,"abstract":"Thanks to a fruitful collaboration between Ifremer (Institut français pour l’exploitation de la mer), Geoazur and CNRS (Centre national de la recherche scientifique), a subsea cabled observatory has been operating offshore Nice (France) since 2015. It is intended to collect measurements of several instruments (seismometer and piezometers) in an area recognized as a natural laboratory for the study of the underlying mechanisms of submarine slope failures. The observatory is made up of two subsea stations, connected to a shore station by an electro-optical cable. All the components of the infrastructure and instruments can be remotely accessed through the internet. The measurements of the instruments are automatically transmitted to data centers. They can be visualized and downloaded on web portals.","PeriodicalId":375793,"journal":{"name":"OCEANS 2019 - Marseille","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121402368","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-06-17DOI: 10.1109/OCEANSE.2019.8867242
Lynda Ben Boudaoud, F. Maussang, R. Garello, Alexis Chevallier
The overarching goal of this paper is to find an automatic bird detection and counting method on aerial images of the ocean. Most of the existing works in the literature are based on heuristic handcrafted feature design, which in most cases affect the effectiveness (the accuracy of classification) and the efficiency (spending much time). In this paper, we propose a method built on a systematic feature learning based classification adopting a new deep Convolutional Neural Network (CNN) architecture. Through this architecture, the feature learning is automated from a multi-dimensional raw input images, by a training step leveraging the JONATHAN dataset via supervised learning. Performances evaluation show that the CNN based architecture classifier achieves an accuracy of 95% on the JONATHAN test data and the overall detection method achieves a classification rate (true positives: birds) of 98% on real images.
{"title":"Marine Bird Detection Based on Deep Learning using High-Resolution Aerial Images","authors":"Lynda Ben Boudaoud, F. Maussang, R. Garello, Alexis Chevallier","doi":"10.1109/OCEANSE.2019.8867242","DOIUrl":"https://doi.org/10.1109/OCEANSE.2019.8867242","url":null,"abstract":"The overarching goal of this paper is to find an automatic bird detection and counting method on aerial images of the ocean. Most of the existing works in the literature are based on heuristic handcrafted feature design, which in most cases affect the effectiveness (the accuracy of classification) and the efficiency (spending much time). In this paper, we propose a method built on a systematic feature learning based classification adopting a new deep Convolutional Neural Network (CNN) architecture. Through this architecture, the feature learning is automated from a multi-dimensional raw input images, by a training step leveraging the JONATHAN dataset via supervised learning. Performances evaluation show that the CNN based architecture classifier achieves an accuracy of 95% on the JONATHAN test data and the overall detection method achieves a classification rate (true positives: birds) of 98% on real images.","PeriodicalId":375793,"journal":{"name":"OCEANS 2019 - Marseille","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128511737","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-06-17DOI: 10.1109/OCEANSE.2019.8867285
F. Menna, E. Nocerino, Mohammad Motasem Nawaf, J. Seinturier, A. Torresani, P. Drap, Fabio Remondino, B. Chemisky
High accuracy underwater inspections are getting more and more important in the underwater industry where time and cost optimization represent nowadays the main innovation drivers. The subsea industry is undergoing a digital transformation process and for this reason, methods that can provide real-time accurate 3D digital measurements are increasingly demanded. This paper provides a short review of the main techniques currently used in subsea metrology to then present an experimental study carried out to evaluate the accuracy potential of three vision-based techniques well-known in photogrammetry, namely visual odometry with and without windowed bundle adjustment, and keyframe based simultaneous localization and mapping (SLAM). The accuracy evaluation is done using an ORUS 3D® subsea photogrammetry system using a certified 3D underwater reference test-field available at COMEX facilities, whose spatial coordinates are known with sub-millimetre accuracy. A critical assessment of results is presented against currently set tolerances for the subsea metrology industry.
{"title":"Towards real-time underwater photogrammetry for subsea metrology applications","authors":"F. Menna, E. Nocerino, Mohammad Motasem Nawaf, J. Seinturier, A. Torresani, P. Drap, Fabio Remondino, B. Chemisky","doi":"10.1109/OCEANSE.2019.8867285","DOIUrl":"https://doi.org/10.1109/OCEANSE.2019.8867285","url":null,"abstract":"High accuracy underwater inspections are getting more and more important in the underwater industry where time and cost optimization represent nowadays the main innovation drivers. The subsea industry is undergoing a digital transformation process and for this reason, methods that can provide real-time accurate 3D digital measurements are increasingly demanded. This paper provides a short review of the main techniques currently used in subsea metrology to then present an experimental study carried out to evaluate the accuracy potential of three vision-based techniques well-known in photogrammetry, namely visual odometry with and without windowed bundle adjustment, and keyframe based simultaneous localization and mapping (SLAM). The accuracy evaluation is done using an ORUS 3D® subsea photogrammetry system using a certified 3D underwater reference test-field available at COMEX facilities, whose spatial coordinates are known with sub-millimetre accuracy. A critical assessment of results is presented against currently set tolerances for the subsea metrology industry.","PeriodicalId":375793,"journal":{"name":"OCEANS 2019 - Marseille","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116701497","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：