The optimization of image processing using GPU


A.S. Kolganov


The approaches to optimize image processing on GPU on the example of the median filtering algorithm are described in the paper.  A comparison is performed with free CPU image processing library, using AVX2 vector instructions. The reached filtration rate of 100 GPixels/sec on 3×3 square  on Titan Pascal GPU and the specific filtration rate of 10.2 GPixels/sec per 1 TFlops on 3×3 square for single accuracy are highest from known ones in the world at present.

Key words

Median filter, image processing, GPU, CUDA.




A.S. Kolganov. The optimization of image processing using GPU // Parallel computing technologies - XII international conference, Pavt ' 2018, Rostov-on-don, 2-6 April 2018 Short articles and poster descriptions, Chelyabinsk: Publishing center SUSU, 2018, P. 266-272