Hybrid Method for Face Description Using LBP and HOG

Authors

  • Dr.Kadhim M. M.AlMousawi
  • Azhaar Abd. Hassan 1Department of Computer Science, College of Education for Pure Science University of Thi-Qar, ThiQar-Iraq
  • Dr.Mustafa J. Hayawi

DOI:

https://doi.org/10.32792/jeps.v10i1.34

Keywords:

Face descriptor, hybrid, LBP, HOG

Abstract

Face recognition has become an important issue in our current life and it is a fundamental task for applications such as face tracking, red – eye removal, face recognition and face expression recognition.
In this paper we present a hybrid approach based on combination of local binary pattern (LBP)and Histogram of Oriented Gradient(HOG). LBP algorithms, which works with the shape and texture information are taken into consideration for representing the facial image. The hog algorithm of descriptor attributes is used to detect the object. It is computed using a dense grid of uniformly spaced cells and uses overlapping local contrast normalization for improved accuracy.
The ORL database was used to test each algorithm for a set of images. The efficiency of the LBP algorithm is evaluated by distinguishing a group of face images from 80%. The HOG algorithm achieved 90% classification accuracy obtained, while the hybrid method 94.25%.

References

Phillips, P., Grother, P., Micheals, R.J., Blackburn, D.M., Tabassi, E., Bone, J.M.: ‟Face recognition

vendor test 2002 results. Technical” report (2003).

Zhao, W., Chellappa, R., Rosenfeld, A., Phillips, P.J.: ‟Face recognition: a liter- ature survey”.

Technical Report CAR-TR-948, Center for Automation Research, University of Maryland (2002).

Phillips, P.J., Wechsler, H., Huang, J., Rauss, P.: ‟The FERET database and evaluation procedure for

face recognition algorithms”. Image and Vision Computing 16 (1998) 295–306

Ojala, T., Pietik¨ainen, M., Harwood, D.: ‟A comparative study of texture measures with classification

based on feature distributions. Pattern Recognition” 29 (1996) 51–59

Ojala, T., Pietik¨ainen, M., M¨aenp¨a¨ a, T.: ‟ Multiresolution gray-scale and rotation invariant texture

classification with local binary patterns”. IEEE Transactions on Pattern Analysis and Machine

Intelligence 24 (2002) 971–987

Dalal N, Triggs B. ‟Histograms of oriented gradients for human detection”. In: IEEE Conference on

Computer Vision and Pattern Recognition (CVPR). San Diego, CA, USA, 2005, 1: 886-893

Downloads

Published

2020-12-02

Issue

Vol. 10 No. 1 (2020)

Section

Articles

License

Copyright Policy

Authors retain copyright of their articles published in the Journal of Education for Pure Science (JEPS).

By submitting their work, authors grant the journal a non-exclusive license to publish, distribute, and archive the article in all formats and media.

License

All articles published in JEPS are licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0).

This license permits unrestricted use, distribution, and reproduction in any medium, provided that the original author(s) and the source are properly credited.

Author Rights

Authors have the right to:

  • Share their articles on personal websites, institutional repositories, and academic platforms

  • Reuse their work in future research and publications

  • Distribute the published version without restriction

Journal Rights

The journal retains the right to:

  • Publish and archive the articles

  • Include them in indexing and archiving systems such as LOCKSS and CLOCKSS

  • Promote and disseminate the published work

Responsibility

The contents of all articles are the sole responsibility of the authors. The journal, editors, and editorial board are not responsible for any errors, opinions, or statements expressed in the published articles.

Open Access Statement

JEPS provides immediate open access to its content, supporting the principle that making research freely available to the public enhances global knowledge exchange.

This work is licensed under a Creative Commons Attribution 4.0 International License.
https://creativecommons.org/licenses/by/4.0/