Integrated Scaling Factor of United Moment Invariant and Rough Sets for Writer Identification

An integrated scaling factor of geometric functions into United Moment Invariant function is proposed for extracting feature of unconstrained words. Conventional regular moments have been proved that there were some drawbacks in the original in terms of scaling factor. Extracted features undergo discretization process before the classification task. In this paper, integrated scaling formulations of Aspect Scaling Invariant are presented in Writer Identification domain to seek the invarianceness of authorship or individuality perseverance, moreover explored the suitability of discretization process for moment invariant data for identify the writer. Mathematical proving and results of computer simulations are included to proof the feasibility of proposed technique in Writer Identification. Keyword: Writer identification, scaling factor, invariancess, moment function.