File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1109/JSEN.2014.2382174
- Scopus: eid_2-s2.0-84926614986
- WOS: WOS:000352085200007
- Find via
Supplementary
- Citations:
- Appears in Collections:
Article: New Object Detection, Tracking, and Recognition Approaches for Video Surveillance Over Camera Network
Title | New Object Detection, Tracking, and Recognition Approaches for Video Surveillance Over Camera Network |
---|---|
Authors | |
Keywords | Bayesian Kalman filter detection recognition tracking Video analytics |
Issue Date | 2015 |
Citation | IEEE Sensors Journal, 2015, v. 15, p. 2679-2691 How to Cite? |
Abstract | Object detection and tracking are two fundamental tasks in multicamera surveillance. This paper proposes a framework for achieving these tasks in a nonoverlapping multiple camera network. A new object detection algorithm using mean shift (MS) segmentation is introduced, and occluded objects are further separated with the help of depth information derived from stereo vision. The detected objects are then tracked by a new object tracking algorithm using a novel Bayesian Kalman filter with simplified Gaussian mixture (BKF-SGM). It employs a Gaussian mixture (GM) representation of the state and noise densities and a novel direct density simplifying algorithm for avoiding the exponential complexity growth of conventional Kalman filters (KFs) using GM. When coupled with an improved MS tracker, a new BKF-SGM with improved MS algorithm with more robust tracking performance is obtained. Furthermore, a nontraining-based object recognition algorithm is employed to support object tracking over nonoverlapping network. Experimental results show that: 1) the proposed object detection algorithm yields improved segmentation results over conventional object detection methods and 2) the proposed tracking algorithm can successfully handle complex scenarios with good performance and low arithmetic complexity. Moreover, the performance of both nontraining- and training-based object recognition algorithms can be improved using our detection and tracking results as input. |
Persistent Identifier | http://hdl.handle.net/10722/238655 |
ISSN | 2023 Impact Factor: 4.3 2023 SCImago Journal Rankings: 1.084 |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | ZHANG, S | - |
dc.contributor.author | WANG, C | - |
dc.contributor.author | Chan, SC | - |
dc.contributor.author | WEI, X | - |
dc.contributor.author | HO, CH | - |
dc.date.accessioned | 2017-02-20T01:24:24Z | - |
dc.date.available | 2017-02-20T01:24:24Z | - |
dc.date.issued | 2015 | - |
dc.identifier.citation | IEEE Sensors Journal, 2015, v. 15, p. 2679-2691 | - |
dc.identifier.issn | 1530-437X | - |
dc.identifier.uri | http://hdl.handle.net/10722/238655 | - |
dc.description.abstract | Object detection and tracking are two fundamental tasks in multicamera surveillance. This paper proposes a framework for achieving these tasks in a nonoverlapping multiple camera network. A new object detection algorithm using mean shift (MS) segmentation is introduced, and occluded objects are further separated with the help of depth information derived from stereo vision. The detected objects are then tracked by a new object tracking algorithm using a novel Bayesian Kalman filter with simplified Gaussian mixture (BKF-SGM). It employs a Gaussian mixture (GM) representation of the state and noise densities and a novel direct density simplifying algorithm for avoiding the exponential complexity growth of conventional Kalman filters (KFs) using GM. When coupled with an improved MS tracker, a new BKF-SGM with improved MS algorithm with more robust tracking performance is obtained. Furthermore, a nontraining-based object recognition algorithm is employed to support object tracking over nonoverlapping network. Experimental results show that: 1) the proposed object detection algorithm yields improved segmentation results over conventional object detection methods and 2) the proposed tracking algorithm can successfully handle complex scenarios with good performance and low arithmetic complexity. Moreover, the performance of both nontraining- and training-based object recognition algorithms can be improved using our detection and tracking results as input. | - |
dc.language | eng | - |
dc.relation.ispartof | IEEE Sensors Journal | - |
dc.subject | Bayesian Kalman filter | - |
dc.subject | detection | - |
dc.subject | recognition | - |
dc.subject | tracking | - |
dc.subject | Video analytics | - |
dc.title | New Object Detection, Tracking, and Recognition Approaches for Video Surveillance Over Camera Network | - |
dc.type | Article | - |
dc.identifier.email | Chan, SC: ascchan@hkucc.hku.hk | - |
dc.identifier.authority | Chan, SC=rp00094 | - |
dc.identifier.doi | 10.1109/JSEN.2014.2382174 | - |
dc.identifier.scopus | eid_2-s2.0-84926614986 | - |
dc.identifier.hkuros | 271378 | - |
dc.identifier.volume | 15 | - |
dc.identifier.spage | 2679 | - |
dc.identifier.epage | 2691 | - |
dc.identifier.eissn | 1558-1748 | - |
dc.identifier.isi | WOS:000352085200007 | - |
dc.identifier.issnl | 1530-437X | - |