Fractal Face Representation and Recognition

Abstract- This paper presents a face representa-tion and recognition scheme based on the theory of fractais. Each face image is represented by its frac-tal model which is a small collection of transforma-tion parameters. The transformation is carried out once for known face images. For recognition, the in-put face image is transformed and its fractal model is then compared against the database of fractal mod-els of known faces. Feedforward neural networks are utilised to implement the compression and recogni-tion parts. Some experimental results are presented. The maximum compression ratio obtained for the suc-cessful recognition of known faces was observed to be 89:l (for a compression threshold of 0.002). I