Tracking consistency metric for video surveillance tracking

The aim of this paper is to propose an additional performance tracking metric. The performance of a tracking algorithm depends on the parameters that are being tracked ranging from reference points that cover edges and corners, to blob parameters such as size, trajectory and color. Based on differen...

Full description

Bibliographic Details
Main Authors: P., Sebastian, V.V., Yap
Format: Conference or Workshop Item
Language:English
Published: 2009
Subjects:
Online Access:http://scholars.utp.edu.my/id/eprint/369/
http://scholars.utp.edu.my/id/eprint/369/1/paper.pdf
_version_ 1848658968197464064
author P., Sebastian
V.V., Yap
author_facet P., Sebastian
V.V., Yap
author_sort P., Sebastian
building UTP Institutional Repository
collection Online Access
description The aim of this paper is to propose an additional performance tracking metric. The performance of a tracking algorithm depends on the parameters that are being tracked ranging from reference points that cover edges and corners, to blob parameters such as size, trajectory and color. Based on different features used in tracking, the tracking metrics range from contingency tables, rates of correct tracks to error differences between tracked features and reference points. Among the tracking metrics that have been developed were tracker detection rate (TDR), object tracking error (OTE) and contingency table. These metrics give a measure of the ability of a tracking algorithm to correctly detect or track a target object. There has been no metric published or utilized to determine the performance of a tracking algorithm in the consistency of maintaining a correct track on the selected target. This paper proposes a method and metric to give a measure of determining the consistency of a tracking. The results of the proposed TC metric have results that range from 0.81 to 0.35 for TDR of 90%, 0.65 to 0.21 for TDR of 80%, 0.5 to 0.16 for TDR of 70%, 0.37 to 0.12 for TDR of 60% and 0.26 to 0.08 for TDR of 50%. © 2009 IEEE.
first_indexed 2025-11-13T07:22:57Z
format Conference or Workshop Item
id oai:scholars.utp.edu.my:369
institution Universiti Teknologi Petronas
institution_category Local University
language English
last_indexed 2025-11-13T07:22:57Z
publishDate 2009
recordtype eprints
repository_type Digital Repository
spelling oai:scholars.utp.edu.my:3692017-01-19T08:25:54Z http://scholars.utp.edu.my/id/eprint/369/ Tracking consistency metric for video surveillance tracking P., Sebastian V.V., Yap TK Electrical engineering. Electronics Nuclear engineering The aim of this paper is to propose an additional performance tracking metric. The performance of a tracking algorithm depends on the parameters that are being tracked ranging from reference points that cover edges and corners, to blob parameters such as size, trajectory and color. Based on different features used in tracking, the tracking metrics range from contingency tables, rates of correct tracks to error differences between tracked features and reference points. Among the tracking metrics that have been developed were tracker detection rate (TDR), object tracking error (OTE) and contingency table. These metrics give a measure of the ability of a tracking algorithm to correctly detect or track a target object. There has been no metric published or utilized to determine the performance of a tracking algorithm in the consistency of maintaining a correct track on the selected target. This paper proposes a method and metric to give a measure of determining the consistency of a tracking. The results of the proposed TC metric have results that range from 0.81 to 0.35 for TDR of 90%, 0.65 to 0.21 for TDR of 80%, 0.5 to 0.16 for TDR of 70%, 0.37 to 0.12 for TDR of 60% and 0.26 to 0.08 for TDR of 50%. © 2009 IEEE. 2009 Conference or Workshop Item NonPeerReviewed application/pdf en http://scholars.utp.edu.my/id/eprint/369/1/paper.pdf P., Sebastian and V.V., Yap (2009) Tracking consistency metric for video surveillance tracking. In: 2009 International Conference on Signal Processing Systems, ICSPS 2009, 15 May 2009 through 17 May 2009, Singapore. http://www.scopus.com/inward/record.url?eid=2-s2.0-70449643397&partnerID=40&md5=e862d7114569a1a935713eb64bb9e385
spellingShingle TK Electrical engineering. Electronics Nuclear engineering
P., Sebastian
V.V., Yap
Tracking consistency metric for video surveillance tracking
title Tracking consistency metric for video surveillance tracking
title_full Tracking consistency metric for video surveillance tracking
title_fullStr Tracking consistency metric for video surveillance tracking
title_full_unstemmed Tracking consistency metric for video surveillance tracking
title_short Tracking consistency metric for video surveillance tracking
title_sort tracking consistency metric for video surveillance tracking
topic TK Electrical engineering. Electronics Nuclear engineering
url http://scholars.utp.edu.my/id/eprint/369/
http://scholars.utp.edu.my/id/eprint/369/
http://scholars.utp.edu.my/id/eprint/369/1/paper.pdf