| Presentation | 2011-01-18 FPGA implementation of human detectin with HOG features and AdaBoost Kazuhiro NEGI, Keisuke DOHI, Yuichiro SHIBATA, Kiyoshi OGURI, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | An increase in in-home accidental deaths of elderly person caused by falling and fainting, which are nonfatal if detected early, has been emerging as a social issue. Camera-based fall detection systems have been widely researched as one of the promising countermeasures of this issue. A process flow of fall detection systems generally consists of two stages; human detection and abnormal action detection, both of which tend to require a large amount of computation. In order to achieve real time detection speed with a small and low-power equipment which can be easily installed in home, efficient implementation as an embedded hardware system is important. In this paper, FPGA implementation of human detection using HOG features and AdaBoost, which is the first step of a fall detection system, is presented. By using binary patterned HOG features, required resources are effectively reduced. As a result of evaluation, our system achieved 60fps of the detection throughput, showing 96.1% and 20.7% of the detection rate and false positive rate, respectively. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | FPGA / HOG fearture / AdaBoost / human detection |
| Paper # | VLD2010-100,CPSY2010-55,RECONF2010-69 |
| Date of Issue |
| Conference Information | |
| Committee | RECONF |
|---|---|
| Conference Date | 2011/1/10(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) | FPGA implementation of human detectin with HOG features and AdaBoost |
| Sub Title (in English) | |
| Keyword(1) | FPGA |
| Keyword(2) | HOG fearture |
| Keyword(3) | AdaBoost |
| Keyword(4) | human detection |
| 1st Author's Name | Kazuhiro NEGI |
| 1st Author's Affiliation | Graduate School of Science and Thechnology, Nagasaki University() |
| 2nd Author's Name | Keisuke DOHI |
| 2nd Author's Affiliation | Graduate School of Science and Thechnology, Nagasaki University |
| 3rd Author's Name | Yuichiro SHIBATA |
| 3rd Author's Affiliation | Nagasaki University |
| 4th Author's Name | Kiyoshi OGURI |
| 4th Author's Affiliation | Nagasaki University |
| Date | 2011-01-18 |
| Paper # | VLD2010-100,CPSY2010-55,RECONF2010-69 |
| Volume (vol) | vol.110 |
| Number (no) | 362 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |