Fingertips tracking based on gradient vector

Ahmad Yahya Dawod, Md. Jan Nordin, Junaidi Abdullah

Research output: Contribution to journalArticle

Abstract

This paper presents a new algorithm to detect and track fingertips position between frames using randomized circle detection. In order to reduce the processing time, a bounding box is created around the fingertips before the algorithm is applied. The randomized circle detection algorithm is employed based on the fact that for a circle, the gradient vector of the edge pixels will always point to the center of the circle. Therefore, if the gradient vector (which is coincident with the gradient line) for all the edge pixels around the circle are drawn, they will intersect at the center of the circle. Hence, the center of the circle can be determined by choosing the point that has the most number of gradient line intersections. The idea can be modified to be applied to semi-circle shape like fingers. Experimental results show good overall performance of individual fingertips tracking with average speed of less than 1s for each frame.

Original languageEnglish
Pages (from-to)165-175
Number of pages11
JournalInternational Journal of Advances in Soft Computing and its Applications
Volume7
Issue number3
Publication statusPublished - 2015

Fingerprint

Pixels
Processing

Keywords

  • Edge detection
  • Fingertips detection
  • Gradient vector
  • Intersection point

ASJC Scopus subject areas

  • Computer Science Applications

Cite this

Fingertips tracking based on gradient vector. / Dawod, Ahmad Yahya; Nordin, Md. Jan; Abdullah, Junaidi.

In: International Journal of Advances in Soft Computing and its Applications, Vol. 7, No. 3, 2015, p. 165-175.

Research output: Contribution to journalArticle

@article{fe302286538c41e481c00b26de07e116,
title = "Fingertips tracking based on gradient vector",
abstract = "This paper presents a new algorithm to detect and track fingertips position between frames using randomized circle detection. In order to reduce the processing time, a bounding box is created around the fingertips before the algorithm is applied. The randomized circle detection algorithm is employed based on the fact that for a circle, the gradient vector of the edge pixels will always point to the center of the circle. Therefore, if the gradient vector (which is coincident with the gradient line) for all the edge pixels around the circle are drawn, they will intersect at the center of the circle. Hence, the center of the circle can be determined by choosing the point that has the most number of gradient line intersections. The idea can be modified to be applied to semi-circle shape like fingers. Experimental results show good overall performance of individual fingertips tracking with average speed of less than 1s for each frame.",
keywords = "Edge detection, Fingertips detection, Gradient vector, Intersection point",
author = "Dawod, {Ahmad Yahya} and Nordin, {Md. Jan} and Junaidi Abdullah",
year = "2015",
language = "English",
volume = "7",
pages = "165--175",
journal = "International Journal of Advances in Soft Computing and its Applications",
issn = "2074-8523",
publisher = "International Center for Scientific Research and Studies (ICSRS)",
number = "3",

}

TY - JOUR

T1 - Fingertips tracking based on gradient vector

AU - Dawod, Ahmad Yahya

AU - Nordin, Md. Jan

AU - Abdullah, Junaidi

PY - 2015

Y1 - 2015

N2 - This paper presents a new algorithm to detect and track fingertips position between frames using randomized circle detection. In order to reduce the processing time, a bounding box is created around the fingertips before the algorithm is applied. The randomized circle detection algorithm is employed based on the fact that for a circle, the gradient vector of the edge pixels will always point to the center of the circle. Therefore, if the gradient vector (which is coincident with the gradient line) for all the edge pixels around the circle are drawn, they will intersect at the center of the circle. Hence, the center of the circle can be determined by choosing the point that has the most number of gradient line intersections. The idea can be modified to be applied to semi-circle shape like fingers. Experimental results show good overall performance of individual fingertips tracking with average speed of less than 1s for each frame.

AB - This paper presents a new algorithm to detect and track fingertips position between frames using randomized circle detection. In order to reduce the processing time, a bounding box is created around the fingertips before the algorithm is applied. The randomized circle detection algorithm is employed based on the fact that for a circle, the gradient vector of the edge pixels will always point to the center of the circle. Therefore, if the gradient vector (which is coincident with the gradient line) for all the edge pixels around the circle are drawn, they will intersect at the center of the circle. Hence, the center of the circle can be determined by choosing the point that has the most number of gradient line intersections. The idea can be modified to be applied to semi-circle shape like fingers. Experimental results show good overall performance of individual fingertips tracking with average speed of less than 1s for each frame.

KW - Edge detection

KW - Fingertips detection

KW - Gradient vector

KW - Intersection point

UR - http://www.scopus.com/inward/record.url?scp=84949761323&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84949761323&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84949761323

VL - 7

SP - 165

EP - 175

JO - International Journal of Advances in Soft Computing and its Applications

JF - International Journal of Advances in Soft Computing and its Applications

SN - 2074-8523

IS - 3

ER -