Improved face recognition by incorporating local color information into the active shape model

Face recognition is one of the most natural biometrics identification approaches. But the accuracy of face recognition is significantly influenced by poses, facial expressions, and lighting conditions. This paper presents a novel robust facial recognition system by incorporating color information in...

Full description

Saved in:
Bibliographic Details
Published in2016 International Conference on Machine Learning and Cybernetics (ICMLC) Vol. 1; pp. 189 - 194
Main Authors Shih-Hsuan Yang, Yu-Xiang Zhan
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.07.2016
Subjects
Active shape model (ASM)
Biometrics
Classification algorithms
Face
Face recognition
Feature extraction
Image color analysis
Training
Online AccessGet full text

Cover

More Information
Summary:Face recognition is one of the most natural biometrics identification approaches. But the accuracy of face recognition is significantly influenced by poses, facial expressions, and lighting conditions. This paper presents a novel robust facial recognition system by incorporating color information into a modified active shape model (ASM). The selected robust geometric and chromatic information of facial landmark points are derived from ASM. Faces are classified using the k-nearest neighbors algorithm with the chosen distance and color features. Compared with the original ASM and its enhanced version, the proposed method consistently achieves a much higher recognition rate at a comparable processing speed.
ISSN:2160-1348
DOI:10.1109/ICMLC.2016.7860899