To provide a similar condition-retrieval system which adequately recognizes a condition of an environment based on information from a plurality of sensors, and a similar condition-retrieval method thereof.
Motion vectors are found at each prescribed time interval (for example, one second) based on acceleration data from sensor nodes 1-1 to 1-4 and are stored in a motion vector DB 23. Every time the motion vectors are found, similarities between the motion vectors and standard motion vectors stored in a standard motion vector DB 24 are found, and observation condition vectors with the similarity to the standard motion vectors are generated and stored with the time (an observation time) in an observation condition DB 26. A similar condition-retrieval module 27 compares the observation condition vectors thus generated with each of the observation condition vectors stored in the observation condition DB 26 and outputs high-similarity observation condition vectors and the observation time thereof as the retrieval result of the similar condition.
COPYRIGHT: (C)2007,JPO&INPIT
Masatsugu Hattori
Tatsumi Yamada
Tsuyoshi Okadome
| JP2005332127A | ||||
| JP7044208A |
平松 薫,外,Pervasive Association:ユビキタス環境における意味統合(2),マルチメディア,分散,協調とモバイル(DICOMO)シンポジウム論文集 1997年~2006年版 Ver.1.1 [DVD-ROM],日本,社団法人情報処理学会,2005年 7月,っp。169-172
服部 正嗣,外,Pervasive Association:ユビキタス環境における意味統合(3),マルチメディア,分散,協調とモバイル(DICOMO)シンポジウム論文集 1997年~2006年版 Ver.1.1 [DVD-ROM],日本,社団法人情報処理学会,2007年 7月,pp.253-256
山田 辰美,外,Pervasive Association:ユビキタス環境における意味統合(4),マルチメディア,分散,協調とモバイル(DICOMO)シンポジウム論文集 1997年~2006年版 Ver.1.1 [DVD-ROM],日本,社団法人情報処理学会,2005年 7月,pp.257-260
Hiroro Kurokawa
Shigeki Yamakawa