A new data compression scheme using orthogonal polynomials

A new orthogonal polynomials based transform coding scheme has been presented in this paper. A class of orthogonal polynomial functions for obtaining polynomial operators with different sizes is proposed. The statistical design of experiments paradigm has been used to separate out the spatial variation within the image region due to discriminable low level features from the spatial variation due to unexplained sources called noise. The degree of this separationability can be controlled by specifying the level of signal-to-noise ratio. The process of separation of signals during polynomial transform coding is based on the proposed statistical hypothesis testing procedures. The results are encouraging