디지털융복합연구 (Journal of Digital Convergence)

  • 제12권11호
  • /
  • Pages.379-386
  • /
  • 2014
  • /
  • 2713-6434(pISSN)
  • /
  • 2713-6442(eISSN)
한국디지털정책학회 (The Society of Digital Policy and Management)
권호 표지
DOI QR코드

DOI QR Code

GPGPU를 이용한 Grabcut의 수행 속도 개선 방법에 관한 연구

A Study of How to Improve Execution Speed of Grabcut Using GPGPU

  • 투고 : 2014.09.18
  • 심사 : 2014.11.20
  • 발행 : 2014.11.28
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 논문에서는 Grabcut 알고리즘의 수행 속도를 효율적으로 개선시키기 위하여 GPU(Graphics Processing Unit)에서 데이터를 처리하는 방법을 제안한다. Grabcut 알고리즘은 뛰어난 성능의 객체 추출 알고리즘으로 기존의 Grabcut 알고리즘은 전경 영역과 배경 영역을 분할한 후 배경 K-클러스터와 전경 K-클러스터로 할당한다. 그리고 할당 된 결과를 점진적으로 개선될 때까지의 과정을 반복한다. 하지만 Grabcut 알고리즘은 반복된 클러스터링 작업으로 인하여 수행 시간이 오래 걸리는 단점이 존재한다. 따라서 GPGPU(General-Purpose computing on Graphics Processing Unit)를 이용해 반복되는 작업을 병렬적으로 처리하여 Grabcut 알고리즘의 수행 속도를 효율적으로 개선시키는 방법을 제안한다. 제안하는 방법으로 Grabcut 알고리즘의 수행시간을 평균 약 90.668% 감소시켰다.

In this paper, the processing speed of Grabcut algorithm in order to efficiently improve the GPU (Graphics Processing Unit) for processing the data from the method. Grabcut algorithm has excellent performance object detection algorithm. Grabcut existing algorithms to split the foreground area and the background area, and then background and foreground K-cluster is assigned a cluster. And assigned to gradually improve the results, until the process is repeated. But Drawback of Grabcut algorithm is the time consumption caused by the repetition of clustering. Thus GPGPU (General-Purpose computing on Graphics Processing Unit) using the repeated operations in parallel by processing Grabcut algorithm to effectively improve the processing speed of the method. We proposed method of execution time of the algorithm reduced the average of about 95.58%.

키워드

참고문헌

  1. Hyun-Ho Han, Gye-Dong Chung, Young-Soo Park, Sang-Hun Lee, Foreground Extraction and Depth Map Creation Method base on Conversion, The Journal of Digital Policy & Management, Vol. 11, No. 1, pp. 243-248, 2013.
  2. Tae-Hoon Yoo, Gang-Seong Lee, Young-Soo Park, Jong-Yong Lee, Sang-Hun Lee, A Study of Depth Estimate using GPGPU in Monocular Image, The Journal of Digital Policy & Manaagement, Vol. 11, No. 12, pp. 345-352, 2013.
  3. Yeong-Kang Lai, Yu-Fan Lai, Ying-Chang Chen, An Effective Hybrid Depth-Generation Algorithm for 2D-to3D Conversion in 3D Displays, Display Technology, Vol. 9, pp. 154-161, 2013. https://doi.org/10.1109/JDT.2012.2224637
  4. Wang Rui, Peng Jinye, Che Liping, Hou Yuting, Improved color image segmentation algorithm base on Grabcut, Applied Mechanics and Materials, Vol. 373-375, pp. 464-467, 2013. https://doi.org/10.4028/www.scientific.net/AMM.373-375.464
  5. Tae-Hee Lee, Bo-Hyun Hwang, Jong-Ho Yun, Myung-Ryul Choi, A Road Extraction Using OpenCV CUDA To Advance The Processing Speed, Journal of digital convergence, Vol. 12, No. 6, pp. 231-236, 2014. https://doi.org/10.14400/JDC.2014.12.6.231
  6. Boykov, Y.Y., Jooly, M. -P., Interactive graph cuts for optimal boundary & region segmentation of objects in N-D images, Computer Vision, ICCV2001, Vol. 1, pp. 105-112, 2001.
  7. Prajapai, H.B., S.K., Analytical Study of Parallel and Distributed Image Processing, Image Information Processing, ICIIO, pp. 1-6, 2011.
  8. Don-Geon Lee, Dong-Kun Shin, The compare performance of CUDA with OpenMP Application for research of GPGPU programming model, IEEK, Vol. 2010, No. 11, pp. 499-500, 2010.
  9. Feng Ji, Heshan Lin, Xiaosong Ma, RSVM: A Region-based Software Virtual Memory for GPU, Parallel Architectures and Compilation Techniques, PACT, pp. 269-278, 2013.
  10. Corporation NVIDIA, CUDA C PROGRAMING GUIDE (version 6.0), NVIDIA Corporation, 2014.
  11. Chen, D., Chen, B., Marnic, G., Fookes, C., Sridharan, S., Improved Grabcut Segmentation via GMM Optimisation, Digital Image Computing : Techniques and Applications, DICTA, pp. 39-45, 2003.
  12. Pother, C., Kolmogorov, V., Blake, A., Grabcut : Interactive Foreground Extraction using Iterated Graph Cuts, ACM Transaction on Graphics (TOG) - Proceedings of ACM SIGGRAPH 2004, TOG, vol. 23, pp. 309-314, 2004.