Face Emotion Identification by Fusing Neural Network and Texture Features: Facial Expression

An emotion describes the feeling of the person. The face is the reflection of the person's emotion. It plays a vital role in communication and expressing the feeling of someone. Face emotion expression make human different from other living beings. In this paper, we have tried to analyze human expression with the combination of neural networks with texture features. Traditionally to extract the face region bounding box is obtained, which contains the nearby information or crop some parts of the face. Therefore to overcome these facial landmarks are used to identify the face region. The proposed approach is divided into three parts. Initially, the face region is extracted; afterward, the face feature vector is obtained using the linear binary pattern to identify the texture feature. Finally, these features are passed in the neural network for the training and testing purpose. It is analyzed that the proposed work performs better as the landmark of the face extracts the specific face region.