One Dimensional Fractal Coder for Online Signature Recognition

Fractal theory has been used for computer graphics, image compression and different fields of pattern recognition. In this paper we simplified a general purpose two-dimensional fractal coder used for image compression. Since in the case of on-line signature recognition, we loose gray levels, contrast and luminosity information, we do not employ these parameters in the fractal coder. Instead, we focused on geometrical relationship between the range block and its best domain block. Then, some features were extracted directly by the proposed one dimensional fractal coder. We show their usefulness in the application of Persian on-line signature recognition