Research on Improved LBP Algorithm Based on Euclidean Distance and Differential Coding

DOI

10.2991/cnci-19.2019.28How to use a DOI?

Keywords

LBP algorithm, feature extraction, recognition rate, Euclidean distance.

Abstract

Local Binary Pattern (LBP) algorithm is a classical algorithm in the field of face recognition. It can capture local detail features, but its robustness and recognition rate are easily affected by external environmental changes.In this paper, an improved LBP algorithm which combines Euclidean distance and differential coding is proposed. The improved EDLBP operator is applied to feature extraction and compared with various improved LBP algorithms in different databases.The experimental results show that the correlation between the EDLBP algorithm and the LBP, MBP, LTP and ELBP algorithms is improved in the databases with illumination diversity and texture rotation. When compared to the recognition rate changes obtained using the CUReT database at different training samples, the highest recognition rate of EDLBP algorithm is 55.49%, 23.36% and 2.46% higher than that of LBP, MBP and LTP algorithms respectively. By comparing similarity of the two face images, the EDLBP algorithm is 1.04%, 2.94%, 4.69%, and 5.56% higher than the latest ELBP algorithm.

Copyright

© 2019, 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  - Xuefei Jia
AU  - Yang Jiao
AU  - Wenhao Zhang
AU  - Junxi Zhao
PY  - 2019/05
DA  - 2019/05
TI  - Research on Improved LBP Algorithm Based on Euclidean Distance and Differential Coding
BT  - Proceedings of the 2019 International Conference on Computer, Network, Communication and Information Systems (CNCI 2019)
PB  - Atlantis Press
SP  - 203
EP  - 207
SN  - 2352-538X
UR  - https://doi.org/10.2991/cnci-19.2019.28
DO  - 10.2991/cnci-19.2019.28
ID  - Jia2019/05
ER  -