DOI QR코드

DOI QR Code

처리 속도 향상을 위해 OpenCV CUDA를 활용한 도로 영역 검출

A Road Region Extraction Using OpenCV CUDA To Advance The Processing Speed

  • 이태희 (한양대학교 전자통신공학과) ;
  • 황보현 (한양대학교 전자전기제어계측공학과) ;
  • 윤종호 (한양대학교 전자통신전파공학과) ;
  • 최명렬 (한양대학교 전자통신공학과)
  • Lee, Tae-Hee (Dept. of Electronics & Communication Engineering, Hanyang University) ;
  • Hwang, Bo-Hyun (Dept. of EECI Engineering, Hanyang University) ;
  • Yun, Jong-Ho (Dept. of Electrical & Computer Engineering, Hanyang University) ;
  • Choi, Myung-Ryul (Dept. of Electronics & Communication Engineering, Hanyang University)
  • 투고 : 2014.03.04
  • 심사 : 2014.06.20
  • 발행 : 2014.06.28

초록

본 논문은 호스트(PC) 기반의 직렬처리 방식으로 도로영역 추출 방식에 디바이스(Graphic Card) 기반의 병렬 처리 방식을 추가함으로써 보다 향상된 처리 속도를 가지는 도로영역검출을 제안하였다. OpenCV CUDA는 기존의 OpenCV와 CUDA를 연동하여 병렬 처리 방식의 많은 함수들을 지원한다. 또한 OpenCV와 CUDA 연동 시 환경 설정이 완료된 OpenCV CUDA 함수들은 사용자의 디바이스(Graphic Card) 사양에 최적화된다. 따라서 OpenCV CUDA 사용은 알고리즘 검증 및 시뮬레이션 결과 도출의 용이성을 제공한다. 제안된 방법은 OpenCV CUDA 와 NVIDIA GeForce GTX 560 Ti 모델의 그래픽 카드를 사용하여 기존 방식보다 3.09배 빠른 처리 속도를 가짐을 실험을 통해 검증한다.

In this paper, we propose a processing speed improvement by adding a parallel processing based on device(graphic card) into a road region extraction by host(PC) based serial processing. The OpenCV CUDA supports the many functions of parallel processing method by interworking a conventional OpenCV with CUDA. Also, when interworking the OpenCV and CUDA, OpenCV functions completed a configuration are optimized the User's device(Graphic Card) specifications. Thus, OpenCV CUDA usage provides an algorithm verification and easiness of simulation result deduction. The proposed method is verified that the proposed method has a about 3.09 times faster processing speed than a conventional method by using OpenCV CUDA and graphic card of NVIDIA GeForce GTX 560 Ti model through experimentation.

키워드

참고문헌

  1. Yongjin Yeom, Yongkuk Cho, "High-Speed Implementations of Block Ciphers on Graphics Processing Units Using CUDA Library", Journal of The Korea Institute of Information Security and Cryptology, Vol. 18, No. 3, pp. 23-32, 2008.
  2. Jun-Chul Kim, Young-Han Jung, Eun-Soo Park, Xuenan Chui, Hak-il Kim, Uk-Youl Huh, "The Implementation of Fast Object Recognition Using Parallel Processing on CPU and GPU", Journal of Institute of Control, Robotics and System, Vol. 15, No. 5, pp. 488-495, 2009. https://doi.org/10.5302/J.ICROS.2009.15.5.488
  3. Kyoung-Hwan Park, Chi-Won Lee, Chang-Woo Lee, "Road Detection using Mean Shift Algorithm and Similarity Region Merging method", Workshop presentatio file, Korea Information Science Society, Vol. 36, No. 4, pp. 437-440, 2009.
  4. Tae-Hee Lee, Bo-Hyun Hwang, Jong-Ho Yun, Byoung-Soo Park, Myung-Ryul Choi, "A Road Extraction Algorithm using Mean-Shift Segmentation and Connected-Component", Journal of Digital Convergence, Vol. 12, no1, pp. 359-364, 2014, 1 https://doi.org/10.14400/JDPM.2014.12.1.359

피인용 문헌

  1. A Study of How to Improve Execution Speed of Grabcut Using GPGPU vol.12, pp.11, 2014, https://doi.org/10.14400/JDC.2014.12.11.379
  2. A Digitalized Recomposition Technique Based on Photo Quality Evaluation Criteria vol.86, pp.1, 2016, https://doi.org/10.1007/s11277-015-2977-y
  3. Photo quality enhancement by relocating subjects vol.19, pp.2, 2016, https://doi.org/10.1007/s10586-016-0547-z
  4. Digital panning shot generator from photographs vol.18, pp.2, 2015, https://doi.org/10.1007/s10586-014-0411-y