Adaptive Classified Vector Quantisation of Non-orthogonal Representations of Images and its Application to Image Compression

A novel digital image compression technique using classified vector quantiser and adaptive transform coding is presented for the efficient representation of still images. Each sub-image is classified into one of five classes based on its directional variances, then adaptively transformed. The transformed sub-image is then vector quantised. The simulation results showed improvements in the peak signal to noise ratio at the expense of increased computational complexity. The improvements in the quality of the compressed images outweigh the computational complexity of the model.