Presentation 2013-05-21
Video based real-time feature extraction and abnormal action detection on an FPGA
Kaoru HAMASAKI, Keisuke DOHI, Yuuichiro SHIBATA, Kiyoshi OGURI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we show an FPGA implementation of video-based real-time feature extraction and abnormal action detection. The abnormal action detection from video image can be applied for monitoring systems for elder people, security systems, and so forth. Real-time abnormal action detection needs high speed feature extraction and to learn abnormal action patterns. We propose an efficient architecture based on streamed processing for principal component analysis. As a result, our FPGA implementation achived 62.5 fps while keeping the same compute accuracy with CPU implementation. Maximum performance achived 8.8 times faster performance compared to CPU implementation with 11 W power consumption.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) FPGA / abnormal action detection / CHLAC / CCIPCA
Paper # RECONF2013-14
Date of Issue

Conference Information
Committee RECONF
Conference Date 2013/5/13(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 Reconfigurable Systems (RECONF)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Video based real-time feature extraction and abnormal action detection on an FPGA
Sub Title (in English)
Keyword(1) FPGA
Keyword(2) abnormal action detection
Keyword(3) CHLAC
Keyword(4) CCIPCA
1st Author's Name Kaoru HAMASAKI
1st Author's Affiliation Nagasaki University()
2nd Author's Name Keisuke DOHI
2nd Author's Affiliation Nagasaki University
3rd Author's Name Yuuichiro SHIBATA
3rd Author's Affiliation Nagasaki University
4th Author's Name Kiyoshi OGURI
4th Author's Affiliation Nagasaki University
Date 2013-05-21
Paper # RECONF2013-14
Volume (vol) vol.113
Number (no) 52
Page pp.pp.-
#Pages 6
Date of Issue