Presented at the International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT2017), Tokat, Turkey, Dec 02, 2017

SETSCI Conference Proceedings, 2017, 1, Page (s): 273-276 , https://doi.org/

Published Date: 08 December 2017    | 1214     7

The use of GPUs for general purpose applications is not a recent approach, but it was quickly becoming widespread with NVIDIA's CUDA (Compute Unified Device Architecture) architecture based on C programming language in 2007. Algorithms suitable for parallel operation, such as image processing applications, can be implemented much more quickly with the GPU. First of all, this paper summarizes the GPU and CUDA. Furthermore, we implement several conventional image processing algorithms on GPU hardware. The CPU and GPU versions of the implemented algorithms will be compared in terms of application speed. For testing purposes, four basic image processing algorithms are implemented using both CPU and GPU.
We have chosen image convolution, histogram equalization, color conversion and median filter. As a result, the GPU version executes much faster than CPU version, especially when the image size is bigger.  

Keywords - Image Processing, GPU, CUDA, Parallel Computing

[1] J. Cheng, M. Grossman and T. McKercher, Professional CUDA C Programming, Elsevier, John Wiley & Sons, Inc., 2014.

[2] I. Tsmots, O. Berezkyi, I. Ihnatiev and I. Gumovska, “Implementation of image processing algorithms based on GPU,” Scientific and Technical Conference, 2016, p. 27-29.

[3] Z. Guler and A. Cinar, "GPU-based image segmentation using level set method with scaling approach," Computer Science & Information Technology (CS & IT), vol. 3, no. 8, pp. 81-92, 2013. doi:10.5121/csit.2013.3808.

[4] J. Sanders and E. Kandrot, CUDA by example: an introduction to general-purpose GPU programming, Addison-Wesley, 2011.

[5] Z. Güler, A. Çinar and E. Özbay, “Investigation of SIFT, SURF, and GPU-SURF Algorithm for Feature Detection,” ICENS International Conference on Engineering and Natural Science, 2016, p. 578-584.

[6] S. Cook, CUDA Programming: A Developer’s Guide to Parallel Computing with GPUs, Elsevier, Morgan Kaufmann, 2012.

[7] NVIDIA, (2017), CUDA C programming guide. [Online]. Available: http://docs.nvidia.com/cuda/pdf/CUDA_C_Programming_Guide.pdf

[8] V. Podlozhnyuk, (2007), Image Convolution with CUDA. [Online]. Available: http://developer.download.nvidia.com/assets/cuda/files/convolutionSe
parable.pdf

[9] (2017) Image programming guide: Performing Convolution Operation, [Online]: https://developer.apple.com/library/content/documentation/Performan
ce/Conceptual/vImage/ConvolutionOperations/ConvolutionOperation s.html.

[10] J. Shajeena and K. Ramar, “A novel way of tracking moving objects in video scenes,” International Conference on Emerging Trends in Electrical and Computer Technology (ICETECT), 2011, p. 805-810.

[11] Z.Yang, Y. Zhu and Y. Pu, “Parallel image processing based on CUDA,” International Conference on Computer Science and Software Engineering, 2008, p.198-201.

[12] A. Deepa and T. Sasipraba, “Age estimation in images using histogram equalization,” 8. International Conference on Advanced Computing (ICoAC), 2017, p.186-190.

[13] G. Rakesh and T. S. Reddy, “YCoCg color image edge detection,” International Journal of Engineering Research and Applications (IJERA), Vol. 2, No. 2, pp. 152-156.

[14] W. Miled and B. Pesquet-Popescu, “The Use of Color Information in Stereo Vision Processing,”. In: High-Quality Visual Experience. Signals and Communication Technology. Springer, 2010.

[15] Y. Said, T. Saidani, and M. Atri, “High-level design for image processing on FPGA using Xilinx AccelDSP,” World Congress on Computer Applications and Information Systems (WCCAIS), 2014.

[16] R. R. Bulyaculov, K. P. Schogoleva, I. N. Yakovlev, and R. A. Roskostov, “Modelling and analysis of the median filter algorithm of suppression of impulse noise,” IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus), 2017, p.649-654.