Title
Multi-gradient features and elongated quinary pattern encoding for image-based facial expression recognition.
Abstract
•A novel texture extraction method for facial expression recognition is proposed.•High performance is attained with both posed and spontaneous facial expression.•Multi-gradient features for magnitude and angle components are derived.•Multi Classifier System based on the multi-class SVM is adopted.•Comparison with the state-of-the-art showed the superiority of the method.
Year
DOI
Venue
2017
10.1016/j.patcog.2017.06.007
Pattern Recognition
Keywords
Field
DocType
Texture feature analysis,Spontaneous facial expression,Multi gradient magnitude and angle images,Elongated quinary pattern,Multi classifier system,Multi-class SVM classifier
Histogram,Computer science,Sobel operator,Artificial intelligence,Classifier (linguistics),Kernel (linear algebra),Computer vision,Pattern recognition,Support vector machine,Feature extraction,Pixel,Machine learning,Encoding (memory)
Journal
Volume
Issue
ISSN
71
1
0031-3203
Citations 
PageRank 
References 
6
0.45
24
Authors
5