Lossless Compression Algorithm for Progressive Transmission of 3-D Medical Image
Yong‐Sung Kim, Whoi-Yul Kim · Journal of Electrical Engineering and Information Science · 1996
This paper presents a new method of lossless predictive coding of 3-D medical image for progressive transmission. Lossless predictive coding may be divided into two consecutive steps: decorrelation step followed by encoding step. In the decorrelation step interpixel redundancies of an image is eliminated. In the encoding step coding redundancies are removed by use of variable length coding like Huffman coding or arithmetic coding. Therefore, the compression ratio of a lossless predictive coding heavily depends upon the decorrelation method. For progressive transmission, hierarchy embedded differential image (REDI) has been extended to deal with 3-D image. Experiments were conducted to verify the performance of 3-D HEDI in terms of the decorrelation efficiency and the progressive transmission efficiency. The former is estimated by the first order entropy and the latter by PSNR at the reconstruction step as a function of the amount of data transmitted. They are compared with those of 2-D HEDI and DPCM. Experimental results indicate that 3-D HEDI outperforms 2-D HEDI and DPCM in beth decorrelation efficiency as well as the progressive transmission efficiency with 3-D medical images.