Quantisation design of multi-dimensional DCT for image coding

This paper presents data folding as a method for rearranging image data in a manner which allows the practical application of the DCT to expand beyond the third dimension. Additionally, the design of the quantisation matrix of a four-dimensional (4-D) DCT based on the human visual system (HVS) and rate-distortion (R-D) theory is presented. Preliminary tests of a hybrid coder utilising 4-D DCT has shown promise.