Abstract
This paper describes a skin color extraction method which has high speed and high accuracy. There were many systems which could achieve the requirements, but they were useless due to tough limitations such as using two cameras to recognize correct coordinates, running only a specific background, wearing several gears, learning the users detailed information in advance, making a personalized database, and so on. In contrast, our approach, which uses the YIQ color system with a background estimation method for skin color extraction, achieved a good performance more than 30 fps, with strong robustness in complicated and poorly illuminated environments.
| Original language | English |
|---|---|
| Pages (from-to) | 210-213 |
| Number of pages | 4 |
| Journal | Artificial Life and Robotics |
| Volume | 12 |
| Issue number | 1-2 |
| DOIs | |
| Publication status | Published - 2008 |
| Externally published | Yes |
Fingerprint
Keywords
- Color system
- Complex background
- Human skin color
- YIQ
ASJC Scopus subject areas
- Artificial Intelligence
- Biochemistry, Genetics and Molecular Biology(all)
Cite this
Color system for skin color extraction. / Tamaki, Emi; Toma, Naruaki.
In: Artificial Life and Robotics, Vol. 12, No. 1-2, 2008, p. 210-213.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Color system for skin color extraction
AU - Tamaki, Emi
AU - Toma, Naruaki
PY - 2008
Y1 - 2008
N2 - This paper describes a skin color extraction method which has high speed and high accuracy. There were many systems which could achieve the requirements, but they were useless due to tough limitations such as using two cameras to recognize correct coordinates, running only a specific background, wearing several gears, learning the users detailed information in advance, making a personalized database, and so on. In contrast, our approach, which uses the YIQ color system with a background estimation method for skin color extraction, achieved a good performance more than 30 fps, with strong robustness in complicated and poorly illuminated environments.
AB - This paper describes a skin color extraction method which has high speed and high accuracy. There were many systems which could achieve the requirements, but they were useless due to tough limitations such as using two cameras to recognize correct coordinates, running only a specific background, wearing several gears, learning the users detailed information in advance, making a personalized database, and so on. In contrast, our approach, which uses the YIQ color system with a background estimation method for skin color extraction, achieved a good performance more than 30 fps, with strong robustness in complicated and poorly illuminated environments.
KW - Color system
KW - Complex background
KW - Human skin color
KW - YIQ
UR - http://www.scopus.com/inward/record.url?scp=73449085860&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=73449085860&partnerID=8YFLogxK
U2 - 10.1007/s10015-007-0469-1
DO - 10.1007/s10015-007-0469-1
M3 - Article
VL - 12
SP - 210
EP - 213
JO - Artificial Life and Robotics
JF - Artificial Life and Robotics
SN - 1433-5298
IS - 1-2
ER -