Performance Evaluation of Magnetic Resonance Images Coding Using Shape Adaptive Discrete Wavelet Transform
G. Suresh, S. Sudha, R. Sukanesh · International Journal of Computer Theory and Engineering · 2009
Region based coding is an important feature in today's image coding techniques as it follows different regions of the image will be encoded at different bit rates and hence at different qualities rather than encoding the entire image with a single quality constraints. This paper evaluates the features of using Shape Adaptive Wavelet Transform for the region based coding of the brain magnetic resonance images, in which the brain part will be encoded with more importance than the background. Shape Adaptive Discrete Wavelet Transform (SA-DWT) can transform the regions of interest and the background on the images independently and the coefficients can be encoded by using the SPIHT coding at different levels. This algorithm was compared with the existing wavelet based coding techniques and a better PSNR was achieved for the same bit rate by reconstructin g the region of interest with high quality than the background. lossless schemes and also maintain the quality of the image. Volumetric medical data are image sequences of any organ for the three dimensional view. These images are correlated both within and across slices to eliminate the redundancy. Some of the existing 3D compression techniques used motion compensation and encoded the image by the mesh formation. The deformation across the slices of the sequence is modeled as motion and the motion compensation algorithm is used to exploit the correlation between the slices (1). This method can provide a lossless compression of the sequence but with increased expense of complexity. Region based compression methods based on wavelet transforms can perform better with relatively less complexity. In JPEG 2000 part - 1 (2), max-shift ROI coding is used which exploits intra band dependency and adopts EBCOT (3) algorithm for coding the bit planes of wavelet coefficients. In this, it is not necessary to send the shape information as side information for decoding but the coefficients will blend with the pixels at the boundary between the ROI and the background. In JPEG 2000 part- 2 (4), scaling based ROI coding is adopted whereby the coefficients associated with the ROI are scaled up by a certain number of bitshift and then the bit planes of the coefficients are encoded plane by plane.