| Presentation | 2008-06-19 Real-time Anomal Movement Detection Based on Position and Movement Direction of People Kyoko SUDO, Hiroyuki OJIMA, Kazuyuki ISO, Kumi JINZENJI, Tatsuya OSAWA, Hideki KOIKE, Hitoshi NAKAZAWA, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | We developed an alogorithm for detecting real-time anomal movement, which we then implemented in a monitoring system. Our algorithm is based on learning two features: the position and the moving direction of people. Using these features to detect anomalies in scenes is advantageous because these features contain more information than those used by conventional systems. Our algorithm adds timing information on movement vectors and extracts the spatial and temporal average of optical flows. After extracting the features, we use an incremental SVM modified for 1-class as the discrimination method to identify anomalies. This enables the online learning to be unsupervised. Experiments showed that when using our system with QVGA-size video as the input, anomal scenes can be detected and that real-time computation is possible. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | security video / anomal detection / optical flow |
| Paper # | DE2008-4,PRMU2008-22 |
| Date of Issue |
| Conference Information | |
| Committee | PRMU |
|---|---|
| Conference Date | 2008/6/12(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) | Real-time Anomal Movement Detection Based on Position and Movement Direction of People |
| Sub Title (in English) | |
| Keyword(1) | security video |
| Keyword(2) | anomal detection |
| Keyword(3) | optical flow |
| 1st Author's Name | Kyoko SUDO |
| 1st Author's Affiliation | NTT Cyber Space Laboratories, NTT Corporation() |
| 2nd Author's Name | Hiroyuki OJIMA |
| 2nd Author's Affiliation | Toyohashi University of Technology |
| 3rd Author's Name | Kazuyuki ISO |
| 3rd Author's Affiliation | NTT Cyber Space Laboratories, NTT Corporation |
| 4th Author's Name | Kumi JINZENJI |
| 4th Author's Affiliation | NTT Cyber Space Laboratories, NTT Corporation |
| 5th Author's Name | Tatsuya OSAWA |
| 5th Author's Affiliation | NTT Cyber Space Laboratories, NTT Corporation |
| 6th Author's Name | Hideki KOIKE |
| 6th Author's Affiliation | NTT Cyber Space Laboratories, NTT Corporation |
| 7th Author's Name | Hitoshi NAKAZAWA |
| 7th Author's Affiliation | NTT Cyber Space Laboratories, NTT Corporation |
| Date | 2008-06-19 |
| Paper # | DE2008-4,PRMU2008-22 |
| Volume (vol) | vol.108 |
| Number (no) | 94 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |