Abnormal Pedestrian Trajectory analysis based on arbitrary-length clustering

Diane Murdock; Jesus Martinez del Rincon

Abnormal Pedestrian Trajectory analysis based on arbitrary-length clustering

Diane Murdock, Jesus Martinez del Rincon

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1217 Downloads (Pure)

Abstract

This papers examines the use of trajectory distance measures and clustering techniques to define normal and abnormal trajectories in the context of pedestrian tracking in public spaces. In order to detect abnormal trajectories, what is meant by a normal trajectory in a given scene is firstly defined. Then every trajectory that deviates from this normality is classified as abnormal. By combining Dynamic Time Warping and a modified K-Means algorithms for arbitrary-length data series, we have developed an algorithm for trajectory clustering and abnormality detection. The final system performs with an overall accuracy of 83% and 75% when tested in two different standard datasets.

Original language	English
Title of host publication	Irish Machine Vision & Image Processing Conference Proceedings 2016
Editors	N. Devaney
Publisher	Irish Pattern Recognition & Classification Society
Pages	17-24
Number of pages	8
ISBN (Print)	978-0-9934207-1-9
Publication status	Published - 26 Aug 2016
Event	18th Irish Machine Vision and Image Processing Conference 2016 - NUIG , Galway, United Kingdom Duration: 24 Aug 2016 → 26 Aug 2016

Conference

Conference	18th Irish Machine Vision and Image Processing Conference 2016
Abbreviated title	IMVIP
Country/Territory	United Kingdom
City	Galway
Period	24/08/2016 → 26/08/2016

Access to Document

Abnormal pedestrian trajectory analysis based on arbitrary-length clustering
Remove embargo on publication. © 2016 The Authors. This is an open access article published under a Creative Commons Attribution-NonCommercial-ShareAlike License (https://creativecommons.org/licenses/by-nc-sa/4.0/), which permits use, distribution and reproduction for non-commercial purposes, provided the author and source are cited and new creations are licensed under the identical terms.
Accepted author manuscript, 1.69 MBLicence: CC BY-NC-SA
Irish Machine Vision and Image Processing Conference Proceedings 2016
©2016 The Authors This work is distributed free of charge by the Irish Pattern Recognition and Classification Society on behalf of the organisers of the Irish Machine Vision and Image Processing Conference, and the contribution authors to this conference. Both organisers and authors own the rights of their contributions to this book. This item is available under the published under a Creative Commons Attribution-NonCommercial-NoDerivs License (https://creativecommons.org/licenses/by-nc-nd/4.0/), which permits distribution and reproduction for non-commercial purposes, provided the author and source are cited.
Final published version, 20 MBLicence: CC BY-NC-ND

https://aran.library.nuigalway.ie/handle/10379/6136Licence: CC BY-NC-ND

Cite this

@inproceedings{aa6689587d4e4938b4f047e5d0feb79d,

title = "Abnormal Pedestrian Trajectory analysis based on arbitrary-length clustering",

abstract = "This papers examines the use of trajectory distance measures and clustering techniques to define normal and abnormal trajectories in the context of pedestrian tracking in public spaces. In order to detect abnormal trajectories, what is meant by a normal trajectory in a given scene is firstly defined. Then every trajectory that deviates from this normality is classified as abnormal. By combining Dynamic Time Warping and a modified K-Means algorithms for arbitrary-length data series, we have developed an algorithm for trajectory clustering and abnormality detection. The final system performs with an overall accuracy of 83% and 75% when tested in two different standard datasets. ",

author = "Diane Murdock and {Martinez del Rincon}, Jesus",

year = "2016",

month = aug,

day = "26",

language = "English",

isbn = "978-0-9934207-1-9",

pages = "17--24",

editor = "N. Devaney",

booktitle = "Irish Machine Vision & Image Processing Conference Proceedings 2016",

publisher = "Irish Pattern Recognition & Classification Society",

note = "18th Irish Machine Vision and Image Processing Conference 2016, IMVIP ; Conference date: 24-08-2016 Through 26-08-2016",

}

Murdock, D & Martinez del Rincon, J 2016, Abnormal Pedestrian Trajectory analysis based on arbitrary-length clustering. in N Devaney (ed.), Irish Machine Vision & Image Processing Conference Proceedings 2016. Irish Pattern Recognition & Classification Society, pp. 17-24, 18th Irish Machine Vision and Image Processing Conference 2016, Galway, United Kingdom, 24/08/2016. <https://aran.library.nuigalway.ie/handle/10379/6136>

TY - GEN

T1 - Abnormal Pedestrian Trajectory analysis based on arbitrary-length clustering

AU - Murdock, Diane

AU - Martinez del Rincon, Jesus

PY - 2016/8/26

Y1 - 2016/8/26

N2 - This papers examines the use of trajectory distance measures and clustering techniques to define normal and abnormal trajectories in the context of pedestrian tracking in public spaces. In order to detect abnormal trajectories, what is meant by a normal trajectory in a given scene is firstly defined. Then every trajectory that deviates from this normality is classified as abnormal. By combining Dynamic Time Warping and a modified K-Means algorithms for arbitrary-length data series, we have developed an algorithm for trajectory clustering and abnormality detection. The final system performs with an overall accuracy of 83% and 75% when tested in two different standard datasets.

AB - This papers examines the use of trajectory distance measures and clustering techniques to define normal and abnormal trajectories in the context of pedestrian tracking in public spaces. In order to detect abnormal trajectories, what is meant by a normal trajectory in a given scene is firstly defined. Then every trajectory that deviates from this normality is classified as abnormal. By combining Dynamic Time Warping and a modified K-Means algorithms for arbitrary-length data series, we have developed an algorithm for trajectory clustering and abnormality detection. The final system performs with an overall accuracy of 83% and 75% when tested in two different standard datasets.

M3 - Conference contribution

SN - 978-0-9934207-1-9

SP - 17

EP - 24

BT - Irish Machine Vision & Image Processing Conference Proceedings 2016

A2 - Devaney, N.

PB - Irish Pattern Recognition & Classification Society

T2 - 18th Irish Machine Vision and Image Processing Conference 2016

Y2 - 24 August 2016 through 26 August 2016

ER -

Abnormal Pedestrian Trajectory analysis based on arbitrary-length clustering

Abstract

Conference

Access to Document

Fingerprint

Cite this