Automatic Countenance Recognition Using DCT-PCA Technique of Facial Patches with High Detection Rate

摘要

Facial expression, being a basic mode of communicating human emotions, finds its applications inhuman-computer interaction (HCI), health-care, surveillance, driver safety, deceit detection etc. Tremendous success being achieved within the fields of face detection and face recognition, emotional computing has received substantial attention among the researchers within the domain of computer vision.Extraction of discriminative features from salient facial patches plays an important role in effective facial features recognition. The accurate detection of facial landmarks improves the localization of the salient patches on face images. This paper proposes a novel framework for expression recognition by using appearance features of chosen facial patches. many distinguished facial patches, depending on the position of facial landmarks, are extracted that are active throughout feeling elicitation. These active patches are further processed to get the salient patches that contain discriminative features for classification of every pair of expressions, thereby choosing totally different facial patches as salient for various pair of expression categories. One-against-one classification technique is adopted using these features. additionally, an automatic learning-free facial landmark detection technique has been proposed, which achieves similar performances as that of other state-of-art landmark detection ways, nevertheless needs considerably less execution time.