A 3-D discrete Hartley transform coder for compression of magnetic resonance images

In this paper, a 3-D discrete Hartley transform coder is proposed to compress 2-D magnetic resonance brain images. The image is segmented into four different sets of subblocks based on the local energy magnitude of the image. Then the subblocks with the same energy are grouped to form a 3-D cuboid. Finally 3-D discrete Hartley transform is applied to the 3-D cuboids and the resulting coefficients are scanned and encoded using Huffman coding. It is found that this technique gives better result compared to the standard JPEG compression method