简介：Multi-beamSonarandSide-scanSonarcompensateeachother.Inordertofullyutilizeallinformation,itisnecessarytofusetwokindsofimageanddata.Andtheimageco-registrationisanimportantandcomplicatedjobbeforefusion.Thispapersuggestscombiningbathymetricdatawithintensityimage,obtainingthecharacteristicpointsthroughtheminimalanglesoflines,andthendecidingthecorrespondingimagepointsbythemaximalcorrelatecoefficientinsearchingspace.Finally,thesecondorderpolynomialisappliedtothedeformationmodel.Aftertheimageshavebeenco-registered,Waveletisusedtofusetheimages.Itisshownthatthisalgorithmcanbeusedintheflatseafloorortheisotropicseabed.Verificationismadeinthepaperwiththeobserveddata.

简介：

简介：

简介：

简介：

简介：Themultibeamsonarscanprovidehydrographicqualitydepthdataaswellasholdthepotentialtoprovidecalibratedmeasurementsoftheseaflooracousticbackscatteringstrength.Therehasbeenmuchinterestinutilizingbackscattersandimagesfrommultibeamsonarforseabedtypeidentificationandmostresultsareobtained.Thispaperhaspresentedafocusedreviewofseveralmainmethodsandrecentdevelopmentsofseafloorclassificationutilizingmultibeamsonardataor/andimages.Theseareincludingthepowerspectralanalysismethods,thetextureanalysis,traditionalBayesianclassificationtheoryandthemostactiveneuralnetworkapproaches.

简介：Sonarimageshavecomplexbackground,lowcontrast,anddeteriorativeedges;thesecharacteristicsmakeitdifficultforresearcherstodisposethesonarobjects.Themulti-resolutionanalysisrepresentsthesignalsindifferentscalesefficiently,whichiswidelyusedinimageprocessing.Waveletsaresuccessfulindisposingpointdiscontinuitiesinonedimension,butnotintwodimensions.ThefiniteRidgelettransform(FRIT)dealsefficientlywiththesingularityinhighdimension.Itpresentsthreeimproveddenoisingapproaches,whicharebasedonFRITandusedinthesonarimagedisposaltechnique.Byexperimentandcomparisonwithtraditionalmethods,theseapproachesnotonlysuppresstheartifacts,butalsoobtaingoodeffectinedgekeepingandSNRofthesonarimagedenoising.

简介：

简介：与健全吸收楔包含一个声纳平台的实验为获得低频率的目的被执行声学的特征。有健全吸收楔的一个声纳站台模型的声学的特征被测量，并且不同的楔的效果在站台上放了区域声学的特征被测试。模型颤动引起的颤动加速和自我噪音处于四个条件被测量：0%，36%，60%，并且当声纳站台在单个频率刺激力量下面时，楔的100%放了区域。一个实验被执行验证相应数字计算。站台区域的数字颤动特征被有限元素方法计算，并且自我噪音在它由颤动引起了被一个经验的公式预言。结论证明数字计算方法罐头部分为获得颤动和健全特征代替试验性的过程。

简介：Anewgray-spatialhistogramisproposed,whichincorporatesspatialinformationwithgraycompositionswithoutsacrificingtherobustnessoftraditionalgrayhistograms.Thepurposeistoconsidertherepresentationroleofgraycompositionsandspatialinformationsimultaneously.Eachentryinthegray-spatialhistogramisthegrayfrequencyandcorrespondingpositioninformationofimages.Intheexperimentsofsonarimagerecognition,theresultsshowthatthegray-spatialhistogramiseffectiveinpracticaluse.

简介：

简介：

简介：这份报纸用生产并且放射的声音为轮船识别和分类建议一个新方法在水下。到那么，一个三步的过程被建议。首先，预处理操作被利用减少噪音效果并且为特征抽取提供信号。第二，一幅二进制图象，用信号分割的频率光谱做了，被形成提取有效特征。第三，一个神经分类器被设计分类信号。二条途径，建议方法和基于分数维图形的方法在真实数据上被比较并且测试。比较结果比基于分数维图形的方法显示了更好的识别能力和建议方法的更柔韧的性能。因此，建议方法能改进识别精确性在水下声学的目标。

简介：Amultiple-inputmultiple-output(MIMO)sonarcansynthesizealarge-aperturevirtualuniformlineararray(ULA)fromasmallnumberofphysicalelements.However,thelargeapertureisobtainedatthecostofagreatnumberofmatchedfilterswithmuchheavycomputationload.Toreducethecomputationload,aMIMOsonarimagingmethodusingavirtualsparselineararray(SLA)isproposed,whichcontainstheofflineandonlineprocessing.Intheofflineprocessing,thevirtualULAoftheMIMOsonaristhinnedtoavirtualSLAbythesimulatedannealingalgorithm,andmatchedfilterscorrespondingtoinactivevirtualelementsareremoved.Intheonlineprocessing,outputsofmatchedfilterscorrespondingtoactiveelementsarecollectedforfurthermultibeamprocessingandhence,thenumberofmatchedfiltersintheechoprocessingprocedureiseffectivelyreduced.Numericalsimulationsshowthattheproposedmethodcanreducethecomputationloadeffectivelywhileobtainingasimilarimagingperformanceasthetraditionalmethod.

简介：Toreducethecomputationburdenofalarge-aperturemultiple-inputmultiple-output(MIMO)sonarimagingsystem,thephase-shiftbeamformer(PSBF)isusedatthecostofbringingtheintensityloss(IL).ThecauseoftheILisanalyzedindetailandavariabletermedasILfactorisdefinedtoquantifythelossamount.TocompensatefortheIL,twomethodstermedasintensitycompensationforthePSBF(IC-PSBF)andthehybridbeamforming(HBF),respectively,areproposed.TheIC-PSBFusespreviouslyestimatedILfactorstocompensateforoutputintensitiesofallPSBFs;andtheHBFappliestheIC-PSBFtothecenterbeamregionandtheshifted-sidebandbeamformer(SSBF)tothesidebeamregion,respectively.Numericalsimulationsdemonstratetheeffectivenessofthetwoproposedmethods.

简介：Withtheincreasingofdetectionabilityofpassivesonar,theweaksignaldetectionprobleminmultipleinterferencesbecomesmoreandmoreimportant.Inthetime/bearingrecord(TBR)displayofsonardetection,whenthereexisttracesofmultipleinterferences,theidentificationofweaksignalisdifficultorimpossible.Theadaptivenoisecancellationtechniqueprovidesthetheoreticalbasisforsuppressingstronginterferences.Butthesolutionforfindingthesteady-stateoptimumfiltermatrixisquitedifficultduetotherealtimecalculationofinversematrixofinputdatacorrelationmatrix.Theiterativeinversebeamforming(IBF)algorithmforsolvingtheoptimumfiltervector,whichisexpressedbyinversematrixoftheoceanenvironmentdata,isderivedinthispaper,bywhich,theoptimumfiltercanbeeventuallyexpressedasasumofseriessimplematricesofconstructedfromsensordata.Basedonthealgorithmproposedinthispaper,someexamplesofatseaexperimentareprovided.Thestronginterferencesarecancelledandtheweaksignalisemerged,evenitdidn'tappearintheconventionalbeamforming(CBF)processing.

简介：一个被动的合成的孔基于为解决港口右舷辨别问题在的阶段修正算法不结盟拖引的成双线的数组声纳，被描述。这个方法通过使用估计的阶段修正进成双线的数组的一个数组创造一个虚拟数组。因为合成虚拟数组在成双线的数组与另外的数组被排列，区别的正确港口右舷结果能被数组基于新合成成双线的数组处理获得。建议方法的效果被显示出由模仿并且在拖引的成双线的数组声纳的海试用结果。与低额外的计算负担，建议方法是容易的适用于实践。

简介：影响的因素在水下车辆声纳自我噪音被分析，并且基因算法和背繁殖(BP)神经网络被联合预言在水下车辆声纳自我噪音。试验性的结果表明那在水下车辆sonarself噪音能被基于的一个GA-BP神经网络精确地预言实际在水下车辆声纳数据。

简介：拖引的站台噪音的干扰特征源于推进器并且拖引身体因为活跃/被动的拖引的数组被分析。在浅水环境，反映的直接波浪和底部/海表面飘动，这被显示出将严重影响声纳系统的性能。为以各种各样的参数计算干扰的到达(DOA)的方向的公式例如数组深度，一些拖引电缆线，是干扰噪音的derived.The效果因为声纳系统的察觉性能是系统模拟的described.The结果为减少这些种干扰的效果提供方法。

简介：小规模的bedforms的词法特征在Dafeng潮汐的公寓上借助于声学的介绍声纳被测量，江苏在2009，并且在Jiulong河口，厦门在2010，分别地。多阀值价值方法被利用揭示bedforms是在场的词法沿着波动。获得的数据集的分析显示出那：(1)沙涟漪能有不规则的形状，并且(2)在bedform形态学的变化在一个单个潮汐的周期以内是小的，但是可能在几个潮汐的周期上重要。分数维图形和海床粗糙的variogram分析揭示了在当前的速度和海床粗糙的分数维的尺寸之间的一种重要关系的存在。作为当前的速度增加，有小规模的底部结构的一个趋势的海床粗糙增加被大规模结构正在代替。而且，大规模底部结构的表面也能由于小规模的特征的缺席变得光滑或由于附加小规模的结构的存在变得更不平。