Robust single object tracker based on kernelled patch of a fixed RGB camera

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

One of the weaknesses of the mean-shift tracker is its limited ability to track a fast-moving object. Not only does the captured image look blurred but the object will often be out of the search window. Moreover, blur issue is a normal occurrence for a fast-moving object by using a commercially available camera due to a low frame rate. This paper aims to develop a robust observation detector for a single object tracker, especially for blurred image, fast-moving and non-rigid object. We propose a fusion of a kernel-based histogram and feature detector based approach. The measurement input is selected from candidate patches that will be generated by matching the vector descriptor. These vectors are built based on the detected points of interest. Two colour spaces are considered where all three channels of RGB are used, while only the hue channel of the HSV space is utilized. The kernel method is employed for better histogram accumulation. The output patch of the previous frame will be the target model where histogram similarity is measured based on Gaussian distribution. The selected output of the patch matching will undergo position smoothing for better precision. Maximum likelihood approach is used to iterate the patch position until the best match is found. The results indicate that our algorithm has a good detection for challenging surroundings and environments. In some cases, our algorithm may have less accuracy but in no case did it failed to detect.

Original languageEnglish
Pages (from-to)1100-1110
Number of pages11
JournalOptik
Volume127
Issue number3
DOIs
Publication statusPublished - 1 Feb 2016

Fingerprint

histograms
Cameras
cameras
Detectors
Gaussian distribution
Maximum likelihood
Fusion reactions
output
detectors
Color
normal density functions
smoothing
fusion
occurrences
color
shift

Keywords

  • Maximum likelihood
  • Neyman-Pearson method
  • Patch matching
  • Robust measurement input
  • Single object tracking

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Electrical and Electronic Engineering
  • Atomic and Molecular Physics, and Optics

Cite this

Robust single object tracker based on kernelled patch of a fixed RGB camera. / Zulkifley, Mohd Asyraf.

In: Optik, Vol. 127, No. 3, 01.02.2016, p. 1100-1110.

Research output: Contribution to journalArticle

@article{3fa86c4e2d5c43a6adb7d0ed2f573dfc,
title = "Robust single object tracker based on kernelled patch of a fixed RGB camera",
abstract = "One of the weaknesses of the mean-shift tracker is its limited ability to track a fast-moving object. Not only does the captured image look blurred but the object will often be out of the search window. Moreover, blur issue is a normal occurrence for a fast-moving object by using a commercially available camera due to a low frame rate. This paper aims to develop a robust observation detector for a single object tracker, especially for blurred image, fast-moving and non-rigid object. We propose a fusion of a kernel-based histogram and feature detector based approach. The measurement input is selected from candidate patches that will be generated by matching the vector descriptor. These vectors are built based on the detected points of interest. Two colour spaces are considered where all three channels of RGB are used, while only the hue channel of the HSV space is utilized. The kernel method is employed for better histogram accumulation. The output patch of the previous frame will be the target model where histogram similarity is measured based on Gaussian distribution. The selected output of the patch matching will undergo position smoothing for better precision. Maximum likelihood approach is used to iterate the patch position until the best match is found. The results indicate that our algorithm has a good detection for challenging surroundings and environments. In some cases, our algorithm may have less accuracy but in no case did it failed to detect.",
keywords = "Maximum likelihood, Neyman-Pearson method, Patch matching, Robust measurement input, Single object tracking",
author = "Zulkifley, {Mohd Asyraf}",
year = "2016",
month = "2",
day = "1",
doi = "10.1016/j.ijleo.2015.10.130",
language = "English",
volume = "127",
pages = "1100--1110",
journal = "Optik",
issn = "0030-4026",
publisher = "Urban und Fischer Verlag Jena",
number = "3",

}

TY - JOUR

T1 - Robust single object tracker based on kernelled patch of a fixed RGB camera

AU - Zulkifley, Mohd Asyraf

PY - 2016/2/1

Y1 - 2016/2/1

N2 - One of the weaknesses of the mean-shift tracker is its limited ability to track a fast-moving object. Not only does the captured image look blurred but the object will often be out of the search window. Moreover, blur issue is a normal occurrence for a fast-moving object by using a commercially available camera due to a low frame rate. This paper aims to develop a robust observation detector for a single object tracker, especially for blurred image, fast-moving and non-rigid object. We propose a fusion of a kernel-based histogram and feature detector based approach. The measurement input is selected from candidate patches that will be generated by matching the vector descriptor. These vectors are built based on the detected points of interest. Two colour spaces are considered where all three channels of RGB are used, while only the hue channel of the HSV space is utilized. The kernel method is employed for better histogram accumulation. The output patch of the previous frame will be the target model where histogram similarity is measured based on Gaussian distribution. The selected output of the patch matching will undergo position smoothing for better precision. Maximum likelihood approach is used to iterate the patch position until the best match is found. The results indicate that our algorithm has a good detection for challenging surroundings and environments. In some cases, our algorithm may have less accuracy but in no case did it failed to detect.

AB - One of the weaknesses of the mean-shift tracker is its limited ability to track a fast-moving object. Not only does the captured image look blurred but the object will often be out of the search window. Moreover, blur issue is a normal occurrence for a fast-moving object by using a commercially available camera due to a low frame rate. This paper aims to develop a robust observation detector for a single object tracker, especially for blurred image, fast-moving and non-rigid object. We propose a fusion of a kernel-based histogram and feature detector based approach. The measurement input is selected from candidate patches that will be generated by matching the vector descriptor. These vectors are built based on the detected points of interest. Two colour spaces are considered where all three channels of RGB are used, while only the hue channel of the HSV space is utilized. The kernel method is employed for better histogram accumulation. The output patch of the previous frame will be the target model where histogram similarity is measured based on Gaussian distribution. The selected output of the patch matching will undergo position smoothing for better precision. Maximum likelihood approach is used to iterate the patch position until the best match is found. The results indicate that our algorithm has a good detection for challenging surroundings and environments. In some cases, our algorithm may have less accuracy but in no case did it failed to detect.

KW - Maximum likelihood

KW - Neyman-Pearson method

KW - Patch matching

KW - Robust measurement input

KW - Single object tracking

UR - http://www.scopus.com/inward/record.url?scp=84959440846&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84959440846&partnerID=8YFLogxK

U2 - 10.1016/j.ijleo.2015.10.130

DO - 10.1016/j.ijleo.2015.10.130

M3 - Article

AN - SCOPUS:84959440846

VL - 127

SP - 1100

EP - 1110

JO - Optik

JF - Optik

SN - 0030-4026

IS - 3

ER -