Facial Expression Recognition Using Difference-Based Local Gradient Pattern
R. Anuja, Emmy Prema C · SSRN Electronic Journal · 2018
In recent years, automatically recognizing facial expression has essential real world applications. It remains a challenging problem and active research topic in computer vision and many novel methods have been proposed to tackle the automatic facial expression recognition problem. One of the difficult issues for a successful facial expression recognition system is to design a robust facial feature descriptor. The proposed method aims to examine the pixel differences on local gradient patterns (LGP) for recognizing facial expression. Two difference-based LGP descriptors are proposed namely angular difference-LGP (AD-LGP), radial difference-LGP (RD-LGP). In facial expression recognition system, two steps are involved namely feature extraction and feature classification. The facial features are extracted in the following three steps: First Gabor Coefficients Maps (GCM) is extracted by using Gabor filters from face images. Second, two difference-based LGP descriptors are applied to GCM to obtain corresponding binary codes. Finally, histogram sequence is described on binary patterns for each region and all histograms are concatenated. K-Nearest Neighbor (K-NN) classifier is used for classification of facial expressions corresponding to different class. The experimental result demonstrates that the recognition rate and the accuracy of the new method are more effective and higher than the other methods under comparison.