Digital image processing using parallel computing based on CUDA technology
Igor P. Skirnevskiy, Andrey Pustovit, M. O. Abdrashitova · Journal of Physics Conference Series · 2017
This article describes expediency of using a graphics processing unit (GPU) in big data processing in the context of digital images processing. It provides a short description of a parallel computing technology and its usage in different areas, definition of the image noise and a brief overview of some noise removal algorithms. It also describes some basic requirements that should be met by certain noise removal algorithm in the projection to computer tomography. It provides comparison of the performance with and without using GPU as well as with different percentage of using CPU and GPU.