Multi-target tracking system using texture

Jun Geun Jeon; Young Ho Kim; Goo Man Park; Kyu Tae Park

doi:10.1117/12.263234

Multi-target tracking system using texture

Jun Geun Jeon, Young Ho Kim, Goo Man Park, Kyu Tae Park

Media IT Engineering Program

Yonsei University

Research output: Contribution to journal › Conference article › peer-review

2 Scopus citations

Abstract

A new method to detect and track non-rigid moving objects against a moving background is presented. In order to discriminate objects from background optical flow based segmentation methods are not appropriate for the case of non-rigid as they have non-uniform optical flow. The proposed method uses optical flow nd texture similarity to discriminate objects from the background. A well-known algorithm from neural networks, Kohonen's self-organizing maps, was adapted to measure the texture similarity.

Original language	English
Pages (from-to)	229-236
Number of pages	8
Journal	Proceedings of SPIE - The International Society for Optical Engineering
Volume	3024
DOIs	https://doi.org/10.1117/12.263234
State	Published - 1997
Event	Visual Communications and Image Processing '97 - San Jose, CA, United States Duration: 12 Feb 1997 → 12 Feb 1997

Keywords

Kohonen maps
Optical flow
Target tracking
Texture

Access to Document

10.1117/12.263234

Cite this

@article{439ea07d53984431acb50a8efb98b62b,

title = "Multi-target tracking system using texture",

abstract = "A new method to detect and track non-rigid moving objects against a moving background is presented. In order to discriminate objects from background optical flow based segmentation methods are not appropriate for the case of non-rigid as they have non-uniform optical flow. The proposed method uses optical flow nd texture similarity to discriminate objects from the background. A well-known algorithm from neural networks, Kohonen's self-organizing maps, was adapted to measure the texture similarity.",

keywords = "Kohonen maps, Optical flow, Target tracking, Texture",

author = "Jeon, \{Jun Geun\} and Kim, \{Young Ho\} and Park, \{Goo Man\} and Park, \{Kyu Tae\}",

year = "1997",

doi = "10.1117/12.263234",

language = "English",

volume = "3024",

pages = "229--236",

journal = "Proceedings of SPIE - The International Society for Optical Engineering",

issn = "0277-786X",

note = "Visual Communications and Image Processing '97 ; Conference date: 12-02-1997 Through 12-02-1997",

}

TY - JOUR

T1 - Multi-target tracking system using texture

AU - Jeon, Jun Geun

AU - Kim, Young Ho

AU - Park, Goo Man

AU - Park, Kyu Tae

PY - 1997

Y1 - 1997

N2 - A new method to detect and track non-rigid moving objects against a moving background is presented. In order to discriminate objects from background optical flow based segmentation methods are not appropriate for the case of non-rigid as they have non-uniform optical flow. The proposed method uses optical flow nd texture similarity to discriminate objects from the background. A well-known algorithm from neural networks, Kohonen's self-organizing maps, was adapted to measure the texture similarity.

AB - A new method to detect and track non-rigid moving objects against a moving background is presented. In order to discriminate objects from background optical flow based segmentation methods are not appropriate for the case of non-rigid as they have non-uniform optical flow. The proposed method uses optical flow nd texture similarity to discriminate objects from the background. A well-known algorithm from neural networks, Kohonen's self-organizing maps, was adapted to measure the texture similarity.

KW - Kohonen maps

KW - Optical flow

KW - Target tracking

KW - Texture

UR - https://www.scopus.com/pages/publications/57749106506

U2 - 10.1117/12.263234

DO - 10.1117/12.263234

M3 - Conference article

AN - SCOPUS:57749106506

SN - 0277-786X

VL - 3024

SP - 229

EP - 236

JO - Proceedings of SPIE - The International Society for Optical Engineering

JF - Proceedings of SPIE - The International Society for Optical Engineering

T2 - Visual Communications and Image Processing '97

Y2 - 12 February 1997 through 12 February 1997

ER -

Multi-target tracking system using texture

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this