robust classification of variable length sonar sequences

robust classification of variable length sonar sequences

ID:34509329

大小:133.85 KB

页数:7页

时间:2019-03-07

robust classification of variable length sonar sequences_第1页
robust classification of variable length sonar sequences_第2页
robust classification of variable length sonar sequences_第3页
robust classification of variable length sonar sequences_第4页
robust classification of variable length sonar sequences_第5页
资源描述:

《robust classification of variable length sonar sequences》由会员上传分享,免费在线阅读,更多相关内容在教育资源-天天文库

1、Robustclassi cationofvariablelengthsonarsequencesJoydeepGhosh,NarsimhamV.GangishettiandSrinivasaV.ChakravarthyDepartmentofElectricalandComputerEngineeringTheUniversityofTexas,Austin,TX78712-1084Abstract.Twotypesofarti cialneuralnetworksareintroducedf

2、ortherobustclassi cationofspatio-temporalsequences.The rstnetworkistheAdaptiveSpatio-TemporalRecognizer(ASTER),whichadaptivelyestimatesthecon dencethata(variablelength)signalofaknownclassispresentbycontinuouslymonitoringasequenceoffeaturevectors.Ifth

3、econ denceforanyclassexceedsathresholdvalueatsomemoment,thesignalisconsideredtobedetectedandclassi ed.ThenonlinearbehaviorofASTERprovidesmorerobustperformancethantherelateddynamictimewarpingalgorithm.ASTERiscomparedwithamorecommonapproachwhereinaself

4、-organizingfeaturemapis rstusedtomapasequenceofextractedfeaturevectorsontoalowerdimensionaltrajectory,whichisthenidenti edusingavariantofthefeedforwardtimedelayneuralnetwork.Theperformanceofthesetwonetworksiscomparedusingarti cialsonogramsaswellasfea

5、turevectorsstringsobtainedfromshort-durationoceanicsignals.1IntroductionManyintelligenttasksinvolvedecision-makingandoutputbehaviorinresponsetospatio-temporalstimuli.Therecognitionandnonlineartransformationofcontinuous-timesignalsorsignalsequencesist

6、husfundamentaltoawiderangeofcognitiveprocesses.Classi cationofspatio-temporalsignalsisalsobasictomanyengineeringapplicationslikespeechrecognition,seismiceventdetection,sonarclassi cationandreal-timecontrol[11,13].Thecentralissueintheprocessingofsuchs

7、ignalsishowpastinputsorhistory"isrepresentedorstored,andhowthishistorya ectstheresponsetothecurrentinputs.Temporalinformationcanberepresentedexplicitlybycreatingaspatial(static)representationoftemporaldata.Tradingtimewithspaceisthebasisfortimedelaye

8、mbeddingstructuresandfrequencyencoding.Alternatively,thepastcanbeimplicitlyrepresentedbyamemoryobtained,forexample,byinsertingtime-delaysintocellsorconnections,usingfeedbackorrecurrentconnections,oraddinginternalstatestotheprocessingcells[13].Thischa

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。