DOI QR코드

DOI QR Code

Implementation of the SIMT based Image Signal Processor for the Image Processing

영상처리를 위한 SIMT 기반 Image Signal Processor 구현

  • Received : 2016.03.02
  • Accepted : 2016.03.21
  • Published : 2016.03.31

Abstract

In this paper, we proposed SIMT based Image Signal Processor which can apply various image preprocessing algorithms and allow parallel processing of application programs such as image recognition. Conventional ISP has the hard-wired image enhancement algorithm of which the processing speed is fast, but there was difficult to optimize performance depending on various image processing algorithms. The proposed ISP improved the processing time applying SIMT architecture and processed a variety of image processing algorithms as an instruction based processor. We used Xilinx Virtex-7 board and the processing time compared to cell multicore processor, ARM Cortex-A9, ARM Cortex-A15 was reduced by about 71 percent, 63 percent and 33 percent, respectively.

본 논문에서는 다양한 영상 전처리 알고리즘들을 적용할 수 있고, 영상 인식과 같이 ISP 응용 프로그램을 병렬로 처리 가능한 SIMT(Single Instruction Multiple Threads) 기반 Image Signal Processor를 제안하였다. 기존의 ISP는 입력 영상의 품질 개선을 위하여 영상 개선 알고리즘이 하드웨어로 설계되어 처리 속도는 빠르지만 다양한 영상 처리 알고리즘에 따라 성능 최적화에 어려움이 있었다. 제안한 ISP는 명령어를 기반으로 한 프로세서로서 다양한 영상 처리 알고리즘을 수행하고 SIMT 구조를 적용하여 알고리즘을 병렬로 수행해 성능을 개선하였다. 제안하는 ISP를 검증하기 위해 Xilinx Virtex-7을 탑재한 VC707 Board를 사용하였으며 cell multicore processor와 비교했을 경우 수행시간이 약 71%, ARM Cortex-A9과 ARM Cortex-A15와 비교하였을 경우 각각 63%, 33% 성능을 개선하였다.

Keywords

References

  1. NVIDIA, "Advanced driver assistance," http://www.nvidia.com
  2. Ho-Sub Yoon, "The Trend of Technology and Market of Image Recognition Service," Communications of the Korea Information Science Society 31(2), pp. 23-31, 2013.
  3. Erik Lindholm, John Nickolls, Stuart Oberman, John Montrym, "NVIDIA Tesla: A Unified Graphics and Computing Architecture," Micro IEEE vol. 28, pp. 39-55, 2008. https://doi.org/10.1109/MM.2008.31
  4. Jeon-yong Lee, "A Design of a Superscalar GP-GPU based on a Stream Processing for Smart Device," The Graduate School of Seokyeong University, 2014.2
  5. Xilinx, "VC707 User Guide," http://www.xilinx.com
  6. J. Barhen, T. Humble, P. Mitra, M. Traweek, "Multi-FFT Vectorization for the Cell Multicore Processor," IEEE/ACM International Conference on Cluster, Cloud and Grid Computing, pp. 780-785, 2010.

Cited by

  1. The parallelization of binarization using a GP-GPU vol.4, pp.4, 2016, https://doi.org/10.17703/IJACT.2016.4.4.57
  2. 슈퍼스칼라 구조를 갖지 않는 고성능 Stream Processor 설계 vol.21, pp.1, 2017, https://doi.org/10.7471/ikeee.2017.21.1.77