Published Paper Details:

In medical field we come across huge amount of images in the forms of scans, reports and so on. Hence, we need to compress the data we come across due to the factor of storage space. Diagnosis of medical image requires detailed analysis of vital portion of the image. In medical image compression, a small loss in the vital portion leads to wrong interpretation. The proposed compression scheme will eliminate this problem by utilizing region based compression which results in the lossless compression of requisite area, where the salient details are stored and lossy compression in the remaining region. In this method, region which contains most required diagnostic features is separated and then encoded without significant loss in diagnostic quality using block based Burrows- Wheeler compression algorithm. Remaining regions are encoded using hybrid fractal encoding algorithm. Finally, both encoded regions are combined together to reconstruct the output image. The performance of the compressed scheme is evaluated in terms of PSNR (Peak Signal to Noise Ratio), CR (Compression Ratio), space saving and time consumption.