Abnormal behavior detection using privacy protected videos

Yumi Iwashita, Shuhei Takaki, Kenichi Morooka, Tokuo Tsuji, Ryo Kurazume

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Intelligent visual surveillance, which relies heavily on human motion detection / recognition and people recognition, has received a lot of attention for its use in effective monitoring of public places. However, there is a concern of loss of privacy due to distinguishable facial and body information. To deal with this issue, researchers proposed to protect privacy example by filtering of face or body areas, and developed methods of people identification from videos in which people's faces has been obfuscated, masked by digital filters. Along the same line of research dealing with videos in which the people faces were masked by filters, this paper introduces a method to detect abnormal behavior. In the proposed method, we first mask face areas in videos by Multiple Instance Learning tracking, and extract silhouette area from each image. We then extract features using affine moment invariants, and perform classification. We build a database including normal and abnormal behaviors, and we show the effectiveness of the proposed method on cases from the database.

Original languageEnglish
Title of host publicationProceedings - 2013 4th International Conference on Emerging Security Technologies, EST 2013
Pages55-57
Number of pages3
DOIs
Publication statusPublished - Dec 1 2013
Event2013 4th International Conference on Emerging Security Technologies, EST 2013 - Cambridge, United Kingdom
Duration: Sep 9 2013Sep 11 2013

Publication series

NameProceedings - 2013 4th International Conference on Emerging Security Technologies, EST 2013

Other

Other2013 4th International Conference on Emerging Security Technologies, EST 2013
CountryUnited Kingdom
CityCambridge
Period9/9/139/11/13

Fingerprint

Digital filters
Masks
Monitoring

All Science Journal Classification (ASJC) codes

  • Software

Cite this

Iwashita, Y., Takaki, S., Morooka, K., Tsuji, T., & Kurazume, R. (2013). Abnormal behavior detection using privacy protected videos. In Proceedings - 2013 4th International Conference on Emerging Security Technologies, EST 2013 (pp. 55-57). [6680186] (Proceedings - 2013 4th International Conference on Emerging Security Technologies, EST 2013). https://doi.org/10.1109/EST.2013.16

Abnormal behavior detection using privacy protected videos. / Iwashita, Yumi; Takaki, Shuhei; Morooka, Kenichi; Tsuji, Tokuo; Kurazume, Ryo.

Proceedings - 2013 4th International Conference on Emerging Security Technologies, EST 2013. 2013. p. 55-57 6680186 (Proceedings - 2013 4th International Conference on Emerging Security Technologies, EST 2013).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Iwashita, Y, Takaki, S, Morooka, K, Tsuji, T & Kurazume, R 2013, Abnormal behavior detection using privacy protected videos. in Proceedings - 2013 4th International Conference on Emerging Security Technologies, EST 2013., 6680186, Proceedings - 2013 4th International Conference on Emerging Security Technologies, EST 2013, pp. 55-57, 2013 4th International Conference on Emerging Security Technologies, EST 2013, Cambridge, United Kingdom, 9/9/13. https://doi.org/10.1109/EST.2013.16
Iwashita Y, Takaki S, Morooka K, Tsuji T, Kurazume R. Abnormal behavior detection using privacy protected videos. In Proceedings - 2013 4th International Conference on Emerging Security Technologies, EST 2013. 2013. p. 55-57. 6680186. (Proceedings - 2013 4th International Conference on Emerging Security Technologies, EST 2013). https://doi.org/10.1109/EST.2013.16
Iwashita, Yumi ; Takaki, Shuhei ; Morooka, Kenichi ; Tsuji, Tokuo ; Kurazume, Ryo. / Abnormal behavior detection using privacy protected videos. Proceedings - 2013 4th International Conference on Emerging Security Technologies, EST 2013. 2013. pp. 55-57 (Proceedings - 2013 4th International Conference on Emerging Security Technologies, EST 2013).
@inproceedings{7134512086fd4201a60053d78d2d85c6,
title = "Abnormal behavior detection using privacy protected videos",
abstract = "Intelligent visual surveillance, which relies heavily on human motion detection / recognition and people recognition, has received a lot of attention for its use in effective monitoring of public places. However, there is a concern of loss of privacy due to distinguishable facial and body information. To deal with this issue, researchers proposed to protect privacy example by filtering of face or body areas, and developed methods of people identification from videos in which people's faces has been obfuscated, masked by digital filters. Along the same line of research dealing with videos in which the people faces were masked by filters, this paper introduces a method to detect abnormal behavior. In the proposed method, we first mask face areas in videos by Multiple Instance Learning tracking, and extract silhouette area from each image. We then extract features using affine moment invariants, and perform classification. We build a database including normal and abnormal behaviors, and we show the effectiveness of the proposed method on cases from the database.",
author = "Yumi Iwashita and Shuhei Takaki and Kenichi Morooka and Tokuo Tsuji and Ryo Kurazume",
year = "2013",
month = "12",
day = "1",
doi = "10.1109/EST.2013.16",
language = "English",
isbn = "9780769550770",
series = "Proceedings - 2013 4th International Conference on Emerging Security Technologies, EST 2013",
pages = "55--57",
booktitle = "Proceedings - 2013 4th International Conference on Emerging Security Technologies, EST 2013",

}

TY - GEN

T1 - Abnormal behavior detection using privacy protected videos

AU - Iwashita, Yumi

AU - Takaki, Shuhei

AU - Morooka, Kenichi

AU - Tsuji, Tokuo

AU - Kurazume, Ryo

PY - 2013/12/1

Y1 - 2013/12/1

N2 - Intelligent visual surveillance, which relies heavily on human motion detection / recognition and people recognition, has received a lot of attention for its use in effective monitoring of public places. However, there is a concern of loss of privacy due to distinguishable facial and body information. To deal with this issue, researchers proposed to protect privacy example by filtering of face or body areas, and developed methods of people identification from videos in which people's faces has been obfuscated, masked by digital filters. Along the same line of research dealing with videos in which the people faces were masked by filters, this paper introduces a method to detect abnormal behavior. In the proposed method, we first mask face areas in videos by Multiple Instance Learning tracking, and extract silhouette area from each image. We then extract features using affine moment invariants, and perform classification. We build a database including normal and abnormal behaviors, and we show the effectiveness of the proposed method on cases from the database.

AB - Intelligent visual surveillance, which relies heavily on human motion detection / recognition and people recognition, has received a lot of attention for its use in effective monitoring of public places. However, there is a concern of loss of privacy due to distinguishable facial and body information. To deal with this issue, researchers proposed to protect privacy example by filtering of face or body areas, and developed methods of people identification from videos in which people's faces has been obfuscated, masked by digital filters. Along the same line of research dealing with videos in which the people faces were masked by filters, this paper introduces a method to detect abnormal behavior. In the proposed method, we first mask face areas in videos by Multiple Instance Learning tracking, and extract silhouette area from each image. We then extract features using affine moment invariants, and perform classification. We build a database including normal and abnormal behaviors, and we show the effectiveness of the proposed method on cases from the database.

UR - http://www.scopus.com/inward/record.url?scp=84893494027&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84893494027&partnerID=8YFLogxK

U2 - 10.1109/EST.2013.16

DO - 10.1109/EST.2013.16

M3 - Conference contribution

AN - SCOPUS:84893494027

SN - 9780769550770

T3 - Proceedings - 2013 4th International Conference on Emerging Security Technologies, EST 2013

SP - 55

EP - 57

BT - Proceedings - 2013 4th International Conference on Emerging Security Technologies, EST 2013

ER -