A hybrid medical image coding based on block truncation coding and residual vector quantisation

The advancement of medical field has witnessed tremendous growth with state of the art imaging technologies for accurate diagnosis, which in turn demands efficient storage of medical images. Further enhance the image quality RVQ is implemented in the proposed method. The proposed work aims at developing an effective algorithm for compressing medical images exploiting the advantages of block truncation coding (BTC) and residual vector quantisation (RVQ). The advantage of the block truncation coding (BTC) is twofold: 1) it is simple to implement; 2) involves less computational complexity. RVQ is used to enhance the quality further. In the proposed method, the input image is compressed using BTC in the first phase. The residual error out of first phase is subjected to RVQ in the second phase. The novel feature of the proposed method is the adaptive procedure of RVQ based on the variance of residual vectors. High variant residual vectors are subject to vector quantisation and low variant vectors to scalar quantisation respectively. Furthermore, the residual values are normalised to positive values, so as to preserve their sign, before quantisation. Experimental results show the superior performance of the proposed method in terms of compression metrics.