Efficient K-Means Fuzzy Cluster Reliability on Angle Oriented Face Recognition

One of the major difficulties in face recognition system is an in-depth pose verification problem. Face recognition is a broad area of research over the last 25 years. However, the field is highly unsolved, largely due to variations in pose, illumination and expression. In this paper, a Ternary based Angle Oriented Face Recognition algorithm is proposed. In this algorithm, the data is made into two clusters namely, Clock wise and Anti-clock wise rotations using Fuzzy method. The image is extracted using angle oriented DCT (Discrete Cosine Transform) that invokes certain normalization techniques. Face Matching is compared to the technique of k-means. Based on Image Recognition pattern, the reliability of clusters is studied using Nelson model. The experimental results are verified.