Facial Expression Recognition Using Local Binary Pattern and Support Vector Machine

Abstract

Facial Expression Recognition (FER) is an important research area in computer vision that aims to identify human emotions from facial images. This paper presents a machine learning approach for FER using Local Binary Pattern (LBP) feature extraction and Support Vector Machine (SVM) classification. Initially, the face image is detected and preprocessed using grayscale conversion, resizing, normalization, and histogram equalization. LBP is then applied to extract local texture features from facial regions such as the eyes, eyebrows, nose, and mouth. The extracted feature histogram is classified using a multiclass SVM. Experimental analysis on CK+, JAFFE, and FER2013 datasets demonstrates that the proposed LBP-SVM approach provides high recognition accuracy with low computational complexity.