activity recognition

activity recognition

ID:40706499

大小:396.71 KB

页数:7页

时间:2019-08-06

activity recognition_第1页
activity recognition_第2页
activity recognition_第3页
activity recognition_第4页
activity recognition_第5页
资源描述:

《activity recognition》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库

1、SmartphonesbasedOnlineActivityRecognitionforIndoorLocalizationusingDeepConvolutionalNeuralNetworkJunYang,KaiCheng,JianfanChen,BaodingZhouandQingquanLiShenzhenKeyLaboratoryofSpatialSmartSensingandServices,ShenzhenUniversityKeyLaboratoryforGeo-EnvironmentMonitoringofCoa

2、stalZoneoftheNationalAdministrationofSurveying,MappingandGeoInformationAbstract—Intheindoorenvironment,theactivityofthepedes-twostepstogetherwithaneuralnetworkabletoautomaticallytriancanreflectsomesemanticinformation,forexample,ifalearnproperfeature.user’sactivityisreco

3、gnizedastakingelevator,thelocationofNowadays,HARhasbeenwidelyusedinourdailylivesintheuserisinferredtobeintheelevator.Theseactivitiescanbeformsofvariousfields.Oneofthemostwidelyusedfieldsusedasthelandmarksforindoorlocalization.Thedevelopmentofsensortechnologyenhancesthesm

4、artphone’ssensingandiselderlymonitoring[1]–[3].Withtherapidincreaseofthecomputationalcapabilities.Usingthesmartphones,theactivityagedpopulationinourcountry,thehealthcareinstitutionsisofthepedestriancanberecognized.Currentmethodsrelyonunderverylargepressureforthehighhea

5、lthcarecostoftheextractingcomplexhand-craftedfeatures,thusleadingtotheelders.HARwhichprovidethemedicalstaffwithmorecrucialincapabilityofrealtimepedestrianactivitiesidentification.Inthisinformationofparticularpatientsforaccidentscangreatlypaper,weproposearealtimepedestri

6、anactivitiesrecognitionmethodbasedondeepconvolutionalneuralnetwork.Anewdeepreducetheworkloadofmedicalinstitutions.Smarthomes[4]convolutionalneural(CNN)networkhasbeendesignedtolearnisanotherdomainsbenefitsfortheHARtechnology.Long-theproperfeaturesautomatically.Experiment

7、sshowthatthetermchronicsleepdisturbancesmaybringmedicalproblemsproposedmethodachievesapproximately95.21%accuracyinlikediabetes,obesity,andpsychologicalconditions,betterabout2secondsinidentifyingninetypesofactivities,includingtoolstoenableaccuratelong-termevaluationofsl

8、eeptimingstill,walk,upstairs,upelevator,upescalator,downelevator,downescalator,downstairsandturning.Besides,wetranspl

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。