http://repository.vnu.edu.vn/handle/VNU_123/10927
This paper proposes a technique to detect abnormalities in the video surveillance based on motion trajectory.
Proposed technique is based on the nature of the route has a certain influence on objects moving on that route, thereby we given a representations routes by segments, combined used Hausdorff distance to calculate the similarity between trajectories.
Therefore, the proposed technique can detect abnormal, even when the object is not complete orbital motion, thus the system can response the video monitoring real-time.
This paper proposes a technique to detect abnormalities in the video surveillance based on motion trajectory.
Proposed technique is based on the nature of the route has a certain influence on objects moving on that route, thereby we given a representations routes by segments, combined used Hausdorff distance to calculate the similarity between trajectories.
Therefore, the proposed technique can detect abnormal, even when the object is not complete orbital motion, thus the system can response the video monitoring real-time.
Title: | An Algorithm to Partition the field of View camera in automatic Surveillance system |
Other Titles: | Một tiếp cận trong phát hiện bất thường dựa vào quỹ đạo trong giám sát video |
Authors: | Ngô Đức Vĩnh, Đỗ Năng Toàn |
Keywords: | Phát hiện bất thường, quỹ đạo, đối tượng chuyển động, video giám sát. |
Issue Date: | 2015 |
Publisher: | ĐHQGHN |
Abstract: | This paper proposes a technique to detect abnormalities in the video surveillance based on motion trajectory. Proposed technique is based on the nature of the route has a certain influence on objects moving on that route, thereby we given a representations routes by segments, combined used Hausdorff distance to calculate the similarity between trajectories. Therefore, the proposed technique can detect abnormal, even when the object is not complete orbital motion, thus the system can response the video monitoring real-time. |
URI: | http://repository.vnu.edu.vn/handle/VNU_123/10927 |
Appears in Collections: | ITI - Papers |
Files in This Item:
Nhận xét
Đăng nhận xét