Institutional Repository
| an inter-modified method of hand tracking and pose recognition under complex background | |
| Yan Yangtian; Zhang Fengjun; Wang Hongan; Dai Guozhong | |
| 2013 | |
| Source | Journal of Computational Information Systems
 |
| ISSN | 1553-9105 |
| Volume | 9Issue:4Pages:1455-1463 |
| English Abstract | This paper proposed an inter-modified method to track and recognize hand under complex background. For more reliable tracking result, two tracking algorithms, Lucas-Kanade tracker and CamShift tracker, are used during the process of tracking. Lucas-Kanade tracker is not only used to track the hand, but its tracking result is also used to modify CamShift tracker. For more reliable recognition result, two tracking results are passed into the recognition module, and then the tracking and recognition inter-modified: combining results of tracking to do recognition can decrease the interference in complex background and increase the efficiency of recognition process; while combining results of recognition to make tracking can eliminate deviations of tracking process. Experiments indicate that compared with traditional tracking method, our method achieves better tracking results. At the same time, our method can provide stable correct hand pose recognition results under complex background. Copyright © 2013 Binary Information Press.; This paper proposed an inter-modified method to track and recognize hand under complex background. For more reliable tracking result, two tracking algorithms, Lucas-Kanade tracker and CamShift tracker, are used during the process of tracking. Lucas-Kanade tracker is not only used to track the hand, but its tracking result is also used to modify CamShift tracker. For more reliable recognition result, two tracking results are passed into the recognition module, and then the tracking and recognition inter-modified: combining results of tracking to do recognition can decrease the interference in complex background and increase the efficiency of recognition process; while combining results of recognition to make tracking can eliminate deviations of tracking process. Experiments indicate that compared with traditional tracking method, our method achieves better tracking results. At the same time, our method can provide stable correct hand pose recognition results under complex background. Copyright © 2013 Binary Information Press. |
| Indexed Type | EI |
| Keyword | Computer Applications Information Systems |
| Department | (1) Institute of Software Chinese Academy of Sciences Beijing 100190 China; (2) University of Chinese Academy of Sciences Beijing 100080 China |
| Language | 英语 |
| Content Type | 期刊论文 |
| URI | http://ir.iscas.ac.cn/handle/311060/15642 |
| Collection | 中国科学院软件研究所 |
| Recommended Citation GB/T 7714 | Yan Yangtian,Zhang Fengjun,Wang Hongan,et al. an inter-modified method of hand tracking and pose recognition under complex background[J]. Journal of Computational Information Systems,2013,9(4):1455-1463. |
| APA | Yan Yangtian,Zhang Fengjun,Wang Hongan,&Dai Guozhong.(2013).an inter-modified method of hand tracking and pose recognition under complex background.Journal of Computational Information Systems,9(4),1455-1463. |
| MLA | Yan Yangtian,et al."an inter-modified method of hand tracking and pose recognition under complex background".Journal of Computational Information Systems 9.4(2013):1455-1463. |
| Files in This Item: | There are no files associated with this item. | |||||
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.
Edit Comment