DefinePK

Abstract

The lossless image compression methods have high demand, particularly in the medical imaging applications. Therefore, we need to increase the image compression acceleration given the demand in the future than the present demand. General Purpose Graphics Processing Unit (GPGPU) is efficient with the development of computing technologies. Therefore, the GPGPU is useful for lossless image compression algorithms to achieve efficiency and speed in performance. This paper is presented with a wavelet tree-based image compression algorithm based on Compute Unified Device Architecture (CUDA) platform. The encoding phases of the compression algorithm have been changed to parallelism and efficiency. With comparable compression ratios, our algorithm works faster than the sequential lossless JPEG-XR and BCWT algorithms. Currently, the parallel BCWT algorithm is subject to further improvement of speed and flexibility.