Hossein Hosseinzadeh

In this thesis, we present a novel approach for recognizing human facial expressions. This method incorporates a feature extraction technique utilizing Local Binary Patterns (LBP) and employs a Radial Basis Function (RBF) network for classification. The parameters of the RBF network are automatically and optimally adjusted to achieve maximum classification accuracy. This process begins with preprocessing and feature extraction from the input vectors using LBP. Subsequently, the extracted features are fed into the RBF network, which classifies the data by solving a linear equation. The results demonstrate the efficiency and effectiveness of the proposed approach. Experiments were conducted on two databases containing various human facial expressions, further validating the method’s performance.