Integer wavelet transform for medical image compression

Abstract

The Integer Wavelet Transform (IWT) is a transform that maps integers to integers and can thus be used for the lossless compression of medical images. A class of IWT which are parametrized simply by one free parameter is presented in this work. The class of IWT is obtained by a lifting based factorization on a class of 6/10 filter pair. By adjusting the free parameter, IWT with different properties and characteristics can be obtained and this feature can be utilized in an adaptive system. Within this class are lifting filters with rational coefficients which are easier to implement than the irrational counterpart.