PoS - Proceedings of Science
Volume 299 - The 7th International Conference on Computer Engineering and Networks (CENet2017) - Session I - Machine Learning
Fast Face Recognition Based on 2D Fractional Fourier Transform
H. Luo*, Y. Wei and M. Zhou
Full text: pdf
Pre-published on: July 17, 2017
Published on: September 06, 2017
Abstract
In order to overcome the deficiency of computational complexity and low accuracy of traditional face recognition algorithm under complex surroundings, this paper proposes a novel face recognition method based on 2D-FrFT. Firstly, we construct Gaussian skin color model to segment potential face region; then adaptive template matching and secondary matching algorithm are adopted to eliminate the impact of background and reduce computational cost in
the process of matching.Meanwhile, LBP operator encodes the
amplitude and phase information of 2D-FrFT as texture feature.
Eventually, a nearest neighbor classifier is employed for classification. Our proposed approach is examined on the public available database. The experimental results demonstrate that th
is method is not only simple and robust, but efficient in
recognition speed and rate.
DOI: https://doi.org/10.22323/1.299.0017
How to cite

Metadata are provided both in "article" format (very similar to INSPIRE) as this helps creating very compact bibliographies which can be beneficial to authors and readers, and in "proceeding" format which is more detailed and complete.

Open Access
Creative Commons LicenseCopyright owned by the author(s) under the term of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.