BROWSE
Detail
Conference Paper
Embedded Face Identification System for a Smart Home
Cited - time in 
Share 
Share
- Authors
- Jang-Hee Yoo, Youngjae Lim, Ho-Sub Yoon, Jae Hong Kim
- Issue Date
- 2017-12
- Citation
- International Conference on Internet (ICONI) 2017 : Symposium, pp.1-2
- Language
- English
- Type
- Conference Paper
- Abstract
- With the evolution of smart home technologies, embedded biometric devices in security systems have gained increasing attention due to the demand for reliable and cost-effective safety and security applications. In this paper, we present the implementation of embedded biometric system that identifies person by using facial images. The LBP-AdaBoost framework is employed for face and eye detection, which is fast and invariant to illumination changes. The face identification algorithm is efficiently enhanced for fixed-point representation and optimized for memory and computational capacity. In experiments, higher performance, which are 95.0% true accept rate with 0.0% false accept rate, have been achieved on real-time test.
- KSP Keywords
- Biometric system, Computational capacity, Facial image, False accept Rate(FAR), Fixed-point, Higher performance, Identification system, Illumination change, Point representation, Real-time test, Safety and security