Automated human emotion detection is a topic of significant interest in the field of computer vision. Over the past\ndecade, much emphasis has been on using facial expression recognition (FER) to extract emotion from facial\nexpressions. Many popular appearance-based methods such as local binary pattern (LBP), local directional pattern\n(LDP) and local ternary pattern (LTP) have been proposed for this task and have been proven both accurate and\nefficient. In recent years, much work has been undertaken into improving these methods. The gradient local ternary\npattern (GLTP) is one such method aimed at increasing robustness to varying illumination and random noise in the\nenvironment. In this paper, GLTP is investigated in more detail and further improvements such as the use of enhanced\npre-processing, a more accurate Scharr gradient operator, dimensionality reduction via principal component analysis\n(PCA) and facial component extraction are proposed. The proposed method was extensively tested on the CK+ and\nJAFFE datasets using a support vector machine (SVM) and shown to further improve the accuracy and efficiency of\nGLTP compared to other common and state-of-the-art methods in literature.
Loading....