IndexStructuresforMultimediaData多媒体数据的索引结构

IndexStructuresforMultimediaData多媒体数据的索引结构

ID:41322741

大小:238.81 KB

页数:10页

时间:2019-08-22

IndexStructuresforMultimediaData多媒体数据的索引结构_第1页
IndexStructuresforMultimediaData多媒体数据的索引结构_第2页
IndexStructuresforMultimediaData多媒体数据的索引结构_第3页
IndexStructuresforMultimediaData多媒体数据的索引结构_第4页
IndexStructuresforMultimediaData多媒体数据的索引结构_第5页
资源描述:

《IndexStructuresforMultimediaData多媒体数据的索引结构》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、IndexStructuresforMultimediaDataFeature-basedApproachJarulojChongstitvatana12301474AdvancedDataStructuresMultimediaDataFeature-basedapproachImage/VoicedataSequencedataGeometricdataTextdescriptorExamplesMovies,musicGenesequenceShape(CAD)DocumentsJarulojChongstitvatana22301474A

2、dvancedDataStructuresQueriesforMultimediaDataPointqueriesGivenadata,findtheexactmatchRangequeriesGivenadata,findsimilardatawithinarangeNearest-neighborqueriesGivenadata,findthemostsimilardataJarulojChongstitvatana32301474AdvancedDataStructuresFeatureTransformationMappingfroma

3、nobjecttoad-dimensionalvector,calledafeaturevector.Whatisthismappingfunction?Forimagedata:colorhistogram,etc.Forsequencedata:numberofeachelementForgeometricdata:slopeofsegmentsofperimeterFortextdescriptor:numberofeachkeywordJarulojChongstitvatana42301474AdvancedDataStructures

4、SimilarityMeasure:distancefunctionGiven2dataobjectsxandy.Let(x,y)bethedistancefunction.(x,y)indicatesthesimilaritybetweendataxandy.Usually(x,y)isbasedonadistancebetweenthefeaturevectorsofxandy.JarulojChongstitvatana52301474AdvancedDataStructuresSimilarityQueriesPointquerie

5、sGivenanobjectx,findanyobjectysuchthat(x,y)=0.RangequeriesGivenanobjectxandathreshold,findanyobjectysuchthat(x,y)<.Nearest-neighborqueriesGivenanobjectx,findanobjectysuchthat(x,y)≤(x,z)foranyobjectzinthedatabase.JarulojChongstitvatana62301474AdvancedDataStructuresDistan

6、ceMeasureEuclideandistanceManhattandistanceMaximumdistanceWeightedEuclideandistanceEllipsoiddistance(x,y)=(i=1,…,d(xi-yi)2)1/2(x,y)=i=1,…,d

7、xi-yi

8、(x,y)=maxi=1,…,d

9、xi-yi

10、(x,y)=(i=1,…,dwi(xi-yi)2)1/2(x,y)=(x-y)TW(x-y)JarulojChongstitvatana72301474AdvancedDataStructuresO

11、therSimilarityQueriesk-Nearest-neighborqueriesGivenanobjectxandanintegerk,findkobjectsy1,y2,…,yk,suchthat,fori=1,2,…,k,(x,yi)≤(x,z)foranyotherobjectzinthedatabase.Approximatenearest-neighborqueriesApproximatek-nearest-neighborqueriesJarulojChongstitvatana82301474AdvancedDat

12、aStructuresRangeQueriesOnk-d-BtreesGridfilesQuadtreesR-treesAlreadyd

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。