Presentation 2007-09-04
Detection of Human Behavior Changing by Surveillance Camera
Tatsuhiko Kagehiro, Takehito Ogata, Hiroshi Sako, Josef Kittler,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We developed the human behavior recognition system by using surveillance camera. This system detects human behavior changing. I adapted the hybrid method which can change the nearest tracking method and particle filter method dynamically, so this system can track some humans who are overlapped. Because of Self Organizing feature Maps, this system can output the classification result without the learning step in advance. In this report, I defined that human walking condition is normal, and the human fighting condition is abnormal. As a result, the rate of false alarm in the walking data set is 1.8%, and the sensitivity rate in the fighting data set is 64.2% in all frames. I found that the most effective feature is the average of the optical flow direction (sine and cosine).
Keyword(in Japanese) (See Japanese page)
Keyword(in English)
Paper # PRMU2007-77,HIP2007-86
Date of Issue

Conference Information
Committee PRMU
Conference Date 2007/8/27(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Detection of Human Behavior Changing by Surveillance Camera
Sub Title (in English)
Keyword(1)
1st Author's Name Tatsuhiko Kagehiro
1st Author's Affiliation Hitachi Research Laboratory, Hitachi, Ltd.()
2nd Author's Name Takehito Ogata
2nd Author's Affiliation Hitachi Research Laboratory, Hitachi, Ltd.
3rd Author's Name Hiroshi Sako
3rd Author's Affiliation Central Research Laboratory, Hitachi, Ltd.
4th Author's Name Josef Kittler
4th Author's Affiliation Centre for Vision, Speech, and Signal Processing, University of Surrey
Date 2007-09-04
Paper # PRMU2007-77,HIP2007-86
Volume (vol) vol.107
Number (no) 206
Page pp.pp.-
#Pages 8
Date of Issue