TY - GEN
T1 - Symbolic representation and retrieval of moving object trajectories
AU - Chen, Lei
AU - Özsu, M. Tamer
AU - Oria, Vincent
PY - 2004
Y1 - 2004
N2 - Searching moving object trajectories of video databases has been applied to many fields, such as video data analysis, content-based video retrieval, video scene classification. In this paper, we propose a novel representation of trajectories, called movement pattern strings, which convert the trajectories into symbolic representations. Movement pattern strings encode both the movement direction and the movement distance information of the trajectories. The distances that are computed in a symbolic space are lower bounds of the distances of original trajectory data, which guarantees that no false dismissals will be introduced using movement pattern strings to retrieve trajectories. In order to improve the retrieval efficiency, we define a modified frequency distance for frequency vectors that are obtained from movement pattern strings to reduce the dimensionality and the computation cost. The experimental results show that using movement pattern strings is almost as effective as using raw trajectories. In addition, the cost of retrieving similar trajectories can greatly be reduced when the modified frequency distance is used as a filter.
AB - Searching moving object trajectories of video databases has been applied to many fields, such as video data analysis, content-based video retrieval, video scene classification. In this paper, we propose a novel representation of trajectories, called movement pattern strings, which convert the trajectories into symbolic representations. Movement pattern strings encode both the movement direction and the movement distance information of the trajectories. The distances that are computed in a symbolic space are lower bounds of the distances of original trajectory data, which guarantees that no false dismissals will be introduced using movement pattern strings to retrieve trajectories. In order to improve the retrieval efficiency, we define a modified frequency distance for frequency vectors that are obtained from movement pattern strings to reduce the dimensionality and the computation cost. The experimental results show that using movement pattern strings is almost as effective as using raw trajectories. In addition, the cost of retrieving similar trajectories can greatly be reduced when the modified frequency distance is used as a filter.
KW - Edit distance on real sequences
KW - Movement pattern string
KW - Symbolic representation
KW - Trajectory
UR - http://www.scopus.com/inward/record.url?scp=15344339663&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=15344339663&partnerID=8YFLogxK
U2 - 10.1145/1026711.1026749
DO - 10.1145/1026711.1026749
M3 - Conference contribution
AN - SCOPUS:15344339663
SN - 1581139403
SN - 9781581139402
T3 - MIR'04 - Proceedings of the 6th ACM SIGMM International Workshop on Multimedia Information Retrieval
SP - 227
EP - 234
BT - MIR'04 - Proceedings of the 6th ACM SIGMM International Workshop on Multimedia Information Retrieval
PB - Association for Computing Machinery (ACM)
T2 - MIR'04 - Proceedings of the 6th ACM SIGMM International Workshop on Multimedia Information Retrieval
Y2 - 15 October 2004 through 16 October 2004
ER -