+高级检索
首页 > 过刊浏览>2023年第6卷第2期 >120-132. DOI:10.19838/j.issn.2096-5753.2023.02.001
PDF HTML阅读 XML下载 导出引用 引用提醒
人工智能辅助的海洋立体观测与探测
作者:
作者简介:

李思源(2001-),男,博士生,主要从事人工智能与海洋信息处理研究

中图分类号:

P714

基金项目:

国家重点研发计划项目资助


Artificial Intelligence-assisted Marine Three-dimensional Observation and Detection
Author:
  • 摘要
    • | |
  • 访问统计
    • |
  • 参考文献 [88]
    • |
  • 相似文献 [20]
    • | | |
  • 文章评论
    摘要:

    海洋立体观测与探测是获取海洋信息的重要手段，是海洋科学研究、环境保护、经济发展的基础。近年来海洋立体观测网络的快速发展带来了观测数据质量与数量的显著提升，进一步推动了海洋信息处理技术从“模型为主”逐步迈向“数据与模型双驱动”的新范式。在这一过程中，人工智能(AI)与海洋信息的交叉融通发挥了重要作用。从海洋立体观测和探测 2 个方面，讨论经典方法的局限性，回顾 AI 辅助下海洋物理场重建、水下目标检测与水下目标定位的研究新进展，重点阐述 AI 辅助的海洋立体观测与探测研究中亟需解决的关键科学问题及潜在的解决思路，并展望了该领域未来的发展方向。

    Abstract:

    Marine three-dimensional observation and detection is an important means to obtain marine information，and is the basis of marine scientific research，environmental protection and economic development. The rapid development of marine three-dimensional observation network in recent years has brought significant improvement in the quality and quantity of observation data，and also promoted the paradigm shift of marine information processing technology from "model-oriented" to "data and model-driven". In this process，the crossintegration of artificial intelligence(AI)and marine information has played an important role. In this paper，we discuss the limitations of classical methods，review new advances in AI-aided marine physical field reconstruction，underwater target detection and localization，highlight the key scientific problems and potential solutions in AI-aided marine three-dimensional observation and detection，and look forward to the future development direction of this field.

    参考文献
    [1] 吴立新,林霄沛,刘永正.“透明海洋”立体观测网构建[J].科学通报,2020,65(25):2654-2661.
    [2] 徐文,李建龙,李一平,等.无人潜水器组网观测探测技术进展与展望[J].前瞻科技,2022,1(2):60-78.
    [3] 沈新蕊,王延辉,杨绍琼,等.水下滑翔机技术发展现状与展望[J].水下无人系统学报,2018,26(2):89-106.
    [4] 王鹏,胡筱敏,熊学军.新型表层漂流浮标体设计分析[J].海洋工程,2017,35(6):125-133.
    [5] WEBB D C,SIMONETTI P J,JONES C P.SLOCUM:an underwater glider propelled by environmentalenergy[J].IEEE Journal of Oceanic Engineering,2001,26(4):447-452.
    [6] ERIKSEN C C,OSSE T J,LIGHT R D,et al.Seaglider:a long-range autonomous underwater vehiclefor oceanographic research[J].IEEE Journal of OceanicEngineering,2001,26(4):424-436.
    [7] WANG X,WANG Y,WANG P,et al.Design,analysis,and testing of Petrel acoustic autonomous underwatervehicle for marine monitoring[J].Physics of Fluids,2022,34(3):37115.
    [8] CACCIA M,BONO R,BRUZZONE G,et al.Unmanned underwater vehicles for scientificapplications and robotics research:the ROMEOproject[J].Marine Technology Society Journal,2000,34(2):3-17.
    [9] 王波,李民,刘世萱,等.海洋资料浮标观测技术应用现状及发展趋势[J].仪器仪表学报,2014,35(11):2654-2661.
    [10] RISER S C,FREELAND H J,ROEMMICH D,et al.Fifteen years of ocean observations with the globalArgo array[J].Nature Climate Change,2016,6(2):145-153.
    [11] GOOS.The global observing system[EB/OL].[2022-12-08].https://www.goosocean.org/.
    [12] ONC.Ocean network Canada[EB/OL].[2022-12-08].https://www.oceannetworks.ca/.
    [13] IOOS.The Integrated Ocean Observing System[EB/OL].[2022-12-08].https://ioos.noaa.gov/.
    [14] OOI.The Ocean Observatories Initiative[EB/OL].[2022-12-08].https://oceanobservatories.org/.
    [15] EMSO.The European Multidisciplinary Seafloor andwater column Observatory[EB/OL].[2022-12-08].https://emso.eu/what-is-emso/.
    [16] MCPHADEN M J.The tropical atmosphere ocean arrayis completed[J].Bulletin of the AmericanMeteorological Society,1995,76(5):739-741.
    [17] 刘帅,陈戈,刘颖洁,等.海洋大数据应用技术分析与趋势研究[J].中国海洋大学学报:自然科学版,2020,50(1):154-164.
    [18] 盛景荃.上海建成中国第一套海底观测组网技术系统[J].华东科技,2009(7):42.
    [19] 陶智.海底观测网络现状与发展分析[J].声学与电子工程,2015(4):45-49.
    [20] 李健,陈荣裕,王盛安,等.国际海洋观测技术发展趋势与中国深海台站建设实践[J].热带海洋学报,2012,31(2):123-133.
    [21] 刘放.摘箬山岛智能海洋观测网信息系统的设计与实现[D].杭州:浙江大学,2016.
    [22] KALININ S V,SUMPTER B G,ARCHIBALD R K.Big-deep-smart data in imaging for guiding materialsdesign[J].Nature materials,2015,14(10):973-980.
    [23] JUMPER J,EVANS R,PRITZEL A,et al.Highlyaccurate protein structure prediction with AlphaFold[J].Nature,2021,596(7873):583-589.
    [24] FAWZI A,BALOG M,HUANG A,et al.Discoveringfaster matrix multiplication algorithms withreinforcement learning[J].Nature,2022,610(7930):47-53.
    [25] KITSIOU D,KARYDIS M.Categorical mapping of marine eutrophication based on ecological indices[J].Science of the Total Environment,2000,255(1-3):113-127.
    [26] RIDGWAY K,DUNN J,WILKIN J.Oceaninterpolation by four-dimensional weighted leastsquares—application to the waters aroundAustralasia[J].Journal of Atmospheric and OceanicTechnology,2002,19(9):1357-1375.
    [27] WEBSTER R,OLIVER M A.Geostatistics forenvironmental scientists[M].New York:John Wiley &Sons,2007.
    [28] CRESSIE N.The origins of Kriging[J].MathematicalGeology,1990,22(3):239-252.
    [29] RASMUSSEN C E.Gaussian processes in machinelearning[C]//Summer School on Machine Learning.Heidelberg:Springer,2003.
    [30] CAVIEDES-NOZAL D,RIIS N A B,HEUCHEL F M,et al.Gaussian processes for sound fieldreconstruction[J].The Journal of the Acoustical Societyof America,2021,149(2):1107-1119.
    [31] YIN F,PAN L,CHEN T,et al.Linear multiplelow-rank kernel based stationary gaussian processesregression for time series[J].IEEE Transactions onSignal Processing,2020,68:5260-5275.
    [32] CHEN W,LI Y,REICH B J,et al.Deepkriging:spatially dependent deep neural networks for spatialprediction[J].arXiv:2007.11972,2020:1-68.
    [33] LI J,HEAP A D.Spatial interpolation methods appliedin the environmental sciences:a review[J].Environmental Modelling & Software,2014,53:173-189.
    [34] LEBLANC L R,MIDDLETON F H.An underwateracoustic sound velocity data model[J].The Journal ofthe Acoustical Society of America,1980,67(6):2055-2062.
    [35] BIANCO M,GERSTOFT P.Dictionary learning ofsound speed profiles[J].The Journal of the AcousticalSociety of America,2017,141(3):1749-1758.
    [36] CHENG L,JI X,ZHAO H,et al.Tensor-based basisfunction learning for three-dimensional sound speedfields[J].The Journal of the Acoustical Society ofAmerica,2022,151(1):269-285.
    [37] CHEN P,CHENG L,ZHANG T,et al.Tensordictionary learning for representing three-dimensionalsound speed fields[J].The Journal of the AcousticalSociety of America,2022,152(5):2601-2616.
    [38] JI X Y,CHENG L,ZHAO H F.Physics-guidedreduced-order representation of three-dimensionalsound speed fields with ocean mesoscale eddies[J].Remote Sensing,2022,14(22):5860.
    [39] EYRE J R,ENGLISH S J,FORSYTHE M.Assimilation of satellite data in numerical weatherprediction.Part I:The early years[J].Quarterly Journalof the Royal Meteorological Society,2020,146(726):49-68.
    [40] BAUER P,THORPE A,BRUNET G.The quietrevolution of numerical weather prediction[J].Nature,2015,525(7567):47-55.
    [41] CHASSIGNET E P,HURLBURT H E,SMEDSTAD OM,et al.The HYCOM(hybrid coordinate ocean model)data assimilative system[J].Journal of MarineSystems,2007,65(1-4):60-83.
    [42] CHEN C,BEARDSLEY R C,COWLES G,et al.Anunstructured-grid,finite-volume community oceanmodel:FVCOM user manual[M].Cambridge,MA,USA:Sea Grant College Program,MassachusettsInstitute of Technology,2012.
    [43] 郑沛楠,宋军,张芳苒,等.常用海洋数值模式简介[J].海洋预报,2008,25(4):108-120.
    [44] WATSON P A G.Applying machine learning toimprove simulations of a chaotic dynamical systemusing empirical error correction[J].Journal ofAdvances in Modeling Earth Systems,2019,11(5):1402-1417.
    [45] TOMPSON J,SCHLACHTER K,SPRECHMANN P,et al.Accelerating Eulerian fluid simulation withconvolutional networks[C]//International Conferenceon Machine Learning.Sydney:IMLS,2017.
    [46] BRANDSTETTER J,WORRALL D,WELLING M.Message passing neural PDE solvers[J].arXiv:2202.03376,2022:1-27.
    [47] YU R,ZHENG S,ANANDKUMAR A,et al.Long-term forecasting using higher order tensorRNNs[J].arXiv:1711.00073,2017:1-24.
    [48] KASHINATH K,MUSTAFA M,ALBERT A,et al.Physics-informed machine learning:case studies forweather and climate modelling[J].PhilosophicalTransactions of the Royal Society A,2021,379(2194):20200093.
    [49] WANG R,WALTERS R,YU R.Meta-learningdynamics forecasting using task inference[J].arXiv:2102.10271,2021:1-20.
    [50] CAMPS-VALLS G,VERRELST J,MUNOZ-MARI J,et al.A survey on Gaussian processes for earth-observation data analysis:a comprehensiveinvestigation[J].IEEE Geoscience and Remote SensingMagazine,2016,4(2):58-78.
    [51] 黄海宁,李宇.水声目标探测技术研究现状与展望[J].中国科学院院刊,2019,34(3):264-271.
    [52] KALYAN B,BALASURIYA A.Sonar based automatictarget detection scheme for underwater environmentsusing CFAR techniques:a comparative study[C]//Proceedings of the 2004 International Symposium onUnderwater Technology.Taipei:IEEE,2004.
    [53] 檀盼龙,吴小兵,张晓宇.基于声呐图像的水下目标识别研究综述[J].数字海洋与水下攻防,2022,5(4):342-353.
    [54] 徐文,鄢社锋,季飞,等.海洋信息获取、传输、处理及融合前沿研究评述[J].中国科学:信息科学,2016,46(8):1053-1085.
    [55] GORMAN R P,SEJNOWSKI T J.Learned classificationof sonar targets using a massively parallel network[J].IEEE Transactions on Acoustics,Speech,and SignalProcessing,1988,36(7):1135-1140.
    [56] WU M H,WANG Q,RIGALL E,et al.ECNet:efficient convolutional networks for side scan sonarimage segmentation[J].Sensors,2019,19(9):2009.
    [57] VALDENEGRO M.Object recognition in forwardlooking sonar images with convolutional neuralnetworks[C]//OCEANS 2016 MTS/IEEE Monterey:IEEE,2016.
    [58] VALDENEGRO M.End-to-end object detection andrecognition in forward-looking sonar images withConvolutional Neural Networks[C]//IEEE/OESAutonomous Underwater Vehicles(AUV).Tokyo:IEEE,2016.
    [59] NGUYEN H,LEE E,BAE C,et al.Multiple objectdetection based on clustering and deep learningmethods[J].Sensors,2020,20(16):4424.
    [60] SEOK J.Active sonar target classification usingmulti-aspect sensing and deep belief networks[J].International Journal of Engineering Research andTechnology,2018,11:1999-2008.
    [61] TERAYAMA K,SHIN K,MIZUNO K,et al.Integration of sonar and optical camera images usingdeep neural network for fish monitoring[J].Aquacultural Engineering,2019,86:1-7.
    [62] LEE S,PARK B,KIM A.Deep learning from shallowdives:sonar image generation and training for underwaterobject detection[J].arXiv:1810.07990,2018:1-8.
    [63] BYUN S H,CHOO Y.Active target classification usinga shallow neural network with dimension reduction[J].The Journal of the Acoustical Society of America,2022,152(4):A62-A63.
    [64] KUBICEK B,GUPTA A S,KIRSTEINS I.Representingsonar target spectral features using a two-dimensionalGabor wavelet[J].The Journal of the AcousticalSociety of America,2020,148(4):2586-2586.
    [65] KUBICEK B,GUPTA A S,KIRSTEINS I.Active sonartarget classification using a physics-cognizant featurerepresentation[J].The Journal of the Acoustical Societyof America,2021,149(4):A55-A56.
    [66] 李启虎.第一讲:进入21世纪的声呐技术[J].物理,2005,35(5):402-407.
    [67] 李启虎,李敏,杨秀庭.水下目标辐射噪声中单频信号分量的检测:理论分析[J].声学学报,2008,33(3):193-196.
    [68] SHIN F B,KIL D H.Full-spectrum signal processingusing a classify-before-detect paradigm[J].The Journalof the Acoustical Society of America,1996,99(4):2188-2197.
    [69] WANG Y,PENG H.Underwater acoustic sourcelocalization using generalized regression neuralnetwork[J].The Journal of the Acoustical Society ofAmerica,2018,143(4):2321-2331.
    [70] SCHMIDT R.Multiple emitter location and signalparameter estimation[J].IEEE Transactions onAntennas and Propagation,1986,34(3):276-280.
    [71] ROY R,KAILATH T.ESPRIT-estimation of signalparameters via rotational invariance techniques[J].IEEE Transactions on Acoustics,Speech,and SignalProcessing,1989,37(7):984-995.
    [72] DONOHO D L.Compressed sensing[J].IEEETransactions on Information Theory,2006,52(4):1289-1306.
    [73] CANDES E J,ROMBERG J,TAO T.Robustuncertainty principles:exact signal reconstruction fromhighly incomplete frequency information[J].IEEETransactions on information theory,2006,52(2):489-509.
    [74] EDELMANN G F,GAUMOND C F.Beamformingusing compressive sensing[J].The Journal of theAcoustical Society of America,2011,130(4):EL232-EL237.
    [75] XENAKI A,GERSTOFT P,MOSEGAARD K.Compressive beamforming[J].The Journal of theAcoustical Society of America,2014,136(1):260-271.
    [76] XENAKI A,GERSTOFT P.Grid-free compressive beamforming[J].The Journal of the Acoustical Societyof America,2015,137(4):1923-1935.
    [77] TIPPING M E.Sparse Bayesian learning and therelevance vector machine[J].Journal of MachineLearning Research,2001,1(6):211-244.
    [78] GERSTOFT P,MECKLENBRAUKER C F,XENAKIA,et al.Multisnapshot sparse Bayesian learning forDOA[J].IEEE Signal Processing Letters,2016,23(10):1469-1473.
    [79] GERSTOFT P,NANNURU S,MECKLENBRAUKERC F,et al.DOA estimation in heteroscedastic noise[J].Signal Processing,2019,161:63-73.
    [80] MA Q,CHENG L,XU W.Experimental verification ofthe minimum Bhattacharyya distance-based sourcebearing estimator[J].JASA Express Letters,2022,2(6):064801.
    [81] MA Q,XU W,ZHOU Y.Statistically robust estimationof source bearing via minimizing the Bhattacharyyadistance[J].The Journal of the Acoustical Society ofAmerica,2022,151(3):1695-1709.
    [82] JENSEN F B,KUPERMAN W A,PORTER M B,et al.Computational ocean acoustics[M].New York:Springer,2011.
    [83] OZARD J M,ZAKARAUSKAS P,KO P.An artificialneural network for range and depth discrimination inmatched field processing[J].The Journal of the AcousticalSociety of America,1991,90(5):2658-2663.
    [84] STEINBERG B Z,BERAN M J,CHIN S H,et al.Aneural network approach to source localization[J].TheJournal of the Acoustical Society of America,1991,90(4):2081-2090.
    [85] NIU H,REEVES E,GERSTOFT P.Source localizationin an ocean waveguide using supervised machinelearning[J].The Journal of the Acoustical Society ofAmerica,2017,142(3):1176-1188.
    [86] WANG Y,PENG H.Underwater acoustic sourcelocalization using generalized regression neuralnetwork[J].The Journal of the Acoustical Society ofAmerica,2018,143(4):2321-2331.
    [87] LIU Y,NIU H,YANG S,et al.Multiple sourcelocalization using learning-based sparse estimation indeep ocean[J].The Journal of the Acoustical Society ofAmerica,2021,150(5):3773-3786.
    [88] OZANICH E,GERSTOFT P,NIU H.A feedforwardneural network for direction-of-arrival estimation[J].The Journal of the Acoustical Society of America,2020,147(3):2035-2048
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

李思源,程磊,张婷,等.人工智能辅助的海洋立体观测与探测[J].数字海洋与水下攻防,2023,6(2):120-132

复制
分享
文章指标
  • 点击次数:314
  • 下载次数: 974
  • HTML阅读次数: 423
  • 引用次数: 0
历史
  • 在线发布日期: 2023-04-24
文章二维码
您是第175910 位访问者
主管单位：中国船舶重工集团有限公司
主办单位：中国船舶重工集团有限公司第七一O研究所
地址：湖北省武汉市江夏区创谷二路710号
传真：027-59312595电话：027-59312586
数字海洋与水下攻防 ® 2025 版权所有