Abstract
This paper presents an automatic facial expression recognition (FER) system. The method proposed is based on an adapted gradient local phase quantization (LPQ) descriptor. Two methods for the quantization of the local phase are proposed here to improve the conventional LPQ. These methods are the “phase thresholding” and the “phase LTP coding”. An experimental study of these methods is performed for identification of both the six and seven basic expressions: happy, surprised, fear, disgust, sad, anger and neutral state. The FER system consists of three main stages. The first step consists of the detection of the face, selection of a region of interest, and normalization of this region. Then extraction of features is done by the adapted gradient LPQ method. The third step is the classification of the emotional states. The SVM are used for this purpose. Evaluation of the system performance is done on the well-known JAFFE, and Cohn and Kanade, databases, with both six and seven facial expressions.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
