Face Description with Monogenic Ternary Pattern: Application to Face Recognition

DOI

10.2991/csic-15.2015.37How to use a DOI?

Keywords

Face recognition, Monogenic signal analysis, Local ternary pattern

Abstract

This paper presents a new face description operator called monogenic ternary pattern (MTP) based on the theory of monogenic signal analysis and local ternary pattern ,then the histogram of MTP (HMTP) is subsequently and applied to the robust face recognition. HMTP consists of two parts:the first part is the monogenic magnitude encoded by local ternary pattern ,and the other is monogenic orientation encoded by quadrant-bit codes. The HMTP is established by concatenating the histograms of MTP of all sub-regions. Extensive experimental results on YALE and FERET face databases show that the proposed MTP algorithm performs much better than state-of-the-art local feature methods.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - He Yan
AU  - Pu Wang
PY  - 2015/07
DA  - 2015/07
TI  - Face Description with Monogenic Ternary Pattern: Application to Face Recognition
BT  - Proceedings of the 2015 International Conference on Computer Science and Intelligent Communication
PB  - Atlantis Press
SP  - 156
EP  - 158
SN  - 2352-538X
UR  - https://doi.org/10.2991/csic-15.2015.37
DO  - 10.2991/csic-15.2015.37
ID  - Yan2015/07
ER  -