Abnormal Event Detection Using Local Sparse Representation

Huamin Ren, Thomas B. Moeslund

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

10 Citations (Scopus)
5 Downloads (Pure)

Abstract

We propose to detect abnormal events via a sparse subspace clustering algorithm. Unlike most existing approaches, which search for optimized normal bases and detect abnormality based on least square error or reconstruction error from the learned normal patterns, we propose an abnormality measurement based on the difference between the normal space and local space. Specifically, we provide a reasonable normal bases through repeated K spectral clustering. Then for each testing feature we first use temporal neighbors to form a local space. An abnormal event is found if any abnormal feature is found that satisfies: the distance between its local space and the normal space is large. We evaluate our method on two public benchmark datasets: UCSD and Subway Entrance datasets. The comparison to the state-of-the-art methods validate our method's effectiveness.
Original languageEnglish
Title of host publication2014 11th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)
Number of pages6
PublisherIEEE Computer Society Press
Publication date2014
Pages125-130
ISBN (Electronic)978-1-4799-4871-0
DOIs
Publication statusPublished - 2014
EventAVSS 2014 - Korea University, Seoul, Korea, Republic of
Duration: 26 Aug 201429 Aug 2014

Conference

ConferenceAVSS 2014
LocationKorea University
CountryKorea, Republic of
CitySeoul
Period26/08/201429/08/2014

Fingerprint

Subways
Clustering algorithms
Testing

Cite this

Ren, H., & Moeslund, T. B. (2014). Abnormal Event Detection Using Local Sparse Representation. In 2014 11th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) (pp. 125-130). IEEE Computer Society Press. https://doi.org/10.1109/AVSS.2014.6918655
Ren, Huamin ; Moeslund, Thomas B. / Abnormal Event Detection Using Local Sparse Representation. 2014 11th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS). IEEE Computer Society Press, 2014. pp. 125-130
@inproceedings{54935ccfda99451dbd5035d783e341c5,
title = "Abnormal Event Detection Using Local Sparse Representation",
abstract = "We propose to detect abnormal events via a sparse subspace clustering algorithm. Unlike most existing approaches, which search for optimized normal bases and detect abnormality based on least square error or reconstruction error from the learned normal patterns, we propose an abnormality measurement based on the difference between the normal space and local space. Specifically, we provide a reasonable normal bases through repeated K spectral clustering. Then for each testing feature we first use temporal neighbors to form a local space. An abnormal event is found if any abnormal feature is found that satisfies: the distance between its local space and the normal space is large. We evaluate our method on two public benchmark datasets: UCSD and Subway Entrance datasets. The comparison to the state-of-the-art methods validate our method's effectiveness.",
author = "Huamin Ren and Moeslund, {Thomas B.}",
year = "2014",
doi = "10.1109/AVSS.2014.6918655",
language = "English",
pages = "125--130",
booktitle = "2014 11th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)",
publisher = "IEEE Computer Society Press",
address = "United States",

}

Ren, H & Moeslund, TB 2014, Abnormal Event Detection Using Local Sparse Representation. in 2014 11th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS). IEEE Computer Society Press, pp. 125-130, AVSS 2014, Seoul, Korea, Republic of, 26/08/2014. https://doi.org/10.1109/AVSS.2014.6918655

Abnormal Event Detection Using Local Sparse Representation. / Ren, Huamin; Moeslund, Thomas B.

2014 11th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS). IEEE Computer Society Press, 2014. p. 125-130.

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - Abnormal Event Detection Using Local Sparse Representation

AU - Ren, Huamin

AU - Moeslund, Thomas B.

PY - 2014

Y1 - 2014

N2 - We propose to detect abnormal events via a sparse subspace clustering algorithm. Unlike most existing approaches, which search for optimized normal bases and detect abnormality based on least square error or reconstruction error from the learned normal patterns, we propose an abnormality measurement based on the difference between the normal space and local space. Specifically, we provide a reasonable normal bases through repeated K spectral clustering. Then for each testing feature we first use temporal neighbors to form a local space. An abnormal event is found if any abnormal feature is found that satisfies: the distance between its local space and the normal space is large. We evaluate our method on two public benchmark datasets: UCSD and Subway Entrance datasets. The comparison to the state-of-the-art methods validate our method's effectiveness.

AB - We propose to detect abnormal events via a sparse subspace clustering algorithm. Unlike most existing approaches, which search for optimized normal bases and detect abnormality based on least square error or reconstruction error from the learned normal patterns, we propose an abnormality measurement based on the difference between the normal space and local space. Specifically, we provide a reasonable normal bases through repeated K spectral clustering. Then for each testing feature we first use temporal neighbors to form a local space. An abnormal event is found if any abnormal feature is found that satisfies: the distance between its local space and the normal space is large. We evaluate our method on two public benchmark datasets: UCSD and Subway Entrance datasets. The comparison to the state-of-the-art methods validate our method's effectiveness.

U2 - 10.1109/AVSS.2014.6918655

DO - 10.1109/AVSS.2014.6918655

M3 - Article in proceeding

SP - 125

EP - 130

BT - 2014 11th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)

PB - IEEE Computer Society Press

ER -

Ren H, Moeslund TB. Abnormal Event Detection Using Local Sparse Representation. In 2014 11th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS). IEEE Computer Society Press. 2014. p. 125-130 https://doi.org/10.1109/AVSS.2014.6918655