Journal of the Korean Institute of Intelligent Systems (한국지능시스템학회논문지)

Korean Institute of Intelligent Systems (한국지능시스템학회)
권호 표지
DOI QR코드

DOI QR Code

Improvement of Disparity Map using Loopy Belief Propagation based on Color and Edge

Disparity 보정을 위한 컬러와 윤곽선 기반 루피 신뢰도 전파 기법

  • Kim, Eun Kyeong (Department of Electrical and Computer Engineering, Pusan National University) ;
  • Cho, Hyunhak (Department of Interdisciplinary Cooperative Course:Robot, Pusan National University) ;
  • Lee, Hansoo (Department of Electrical and Computer Engineering, Pusan National University) ;
  • Wibowo, Suryo Adhi (Department of Electrical and Computer Engineering, Pusan National University) ;
  • Kim, Sungshin (School of Electrical and Computer Engineering, Pusan National University)
  • Received : 2015.09.08
  • Accepted : 2015.10.01
  • Published : 2015.10.25
Download PDF
⟨ Previous Next ⟩

Abstract

Stereo images have an advantage of calculating depth(distance) values which can not analyze from 2-D images. However, depth information obtained by stereo images has due to following reasons: it can be obtained by computation process; mismatching occurs when stereo matching is processing in occlusion which has an effect on accuracy of calculating depth information. Also, if global method is used for stereo matching, it needs a lot of computation. Therefore, this paper proposes the method obtaining disparity map which can reduce computation time and has higher accuracy than established method. Edge extraction which is image segmentation based on feature is used for improving accuracy and reducing computation time. Color K-Means method which is image segmentation based on color estimates correlation of objects in an image. And it extracts region of interest for applying Loopy Belief Propagation(LBP). For this, disparity map can be compensated by considering correlation of objects in the image. And it can reduce computation time because of calculating region of interest not all pixels. As a result, disparity map has more accurate and the proposed method reduces computation time.

스테레오 영상은 2-D 영상으로 분석할 수 없는 깊이(거리) 정보를 포함하고 있다. 하지만 연산을 통해서 거리정보를 얻을 수 있기 때문에 계산 값의 신뢰도가 낮고, 폐색된 공간 등의 영향으로 오차가 발생한다. 또한 Stereo Matching 시 Global Method를 사용할 경우, 많은 연산량에 따라 계산 시간이 오래 걸린다. 따라서 본 논문에서는 연산 시간이 짧고 더 높은 정확도를 갖는 Disparity Map을 구하는 방법을 제안한다. 특징 기반 영상분할 기법인 윤곽선 추출을 통해 정확도는 높이고 연산 시간은 줄였다. 컬러 기반 영상 분할 기법인 Color K-Means를 통해 관심 영역을 추출하고, 이를 기반으로 Loopy Belief Propagation(LBP)을 접목하였다. 제안하는 방법을 적용함으로 영상 내 물체들의 연관성을 고려한 보정이 가능하였고, 관심 영역 추출에 따라 연산 시간을 줄일 수 있었다. 실험 결과, 기존의 방법들보다 연산 시간이 짧고 정확도가 높은 Disparity Map을 얻을 수 있었다.

Keywords

References

  1. E. T. Baumgartner and S. B. Skaar, "An Autonomous Vision-Based Mobile Robot," IEEE Transactions on Automatic Control, Vol. 39, No. 3, pp. 493-502, 1994. https://doi.org/10.1109/9.280748
  2. C. H. Han and K. B. Sim, "Real-Time Mapping of Mobile Robot on Stereo Vision," Journal of Korean Institute of Intelligent Systems, Vol. 20, No. 1, pp. 60-65, 2010. https://doi.org/10.5391/JKIIS.2010.20.1.060
  3. H. H. Min, D. S. Yoo and Y. T. Kim, "Fuzzy Tracking Control Based on Stereo Images for Tracking of Moving Robot," Journal of Korean Institute of Intelligent Systems, Vol. 22, No. 2, pp. 198-204, 2012. https://doi.org/10.5391/JKIIS.2012.22.2.198
  4. I. Oh, Computer Vision, Hanbit Academy Inc., 2014.
  5. D. Scharstein and R. Szeliski, "A Taxonomy and Evaluation of Dense Two-Frame Stereo Correspondence Algorithms," International Journal of Computer Vision, Vol. 47, No. 7, pp. 7-42, 2002. https://doi.org/10.1023/A:1014573219977
  6. H. Hirschmuller, "Accurate and Efficient Stereo Processing by Semi-Global Matching and Mutual Information," IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Vol. 2, pp. 807-814, 2005.
  7. P. F. Felzenszwalb and D. P. Huttenlocher, "Efficient Belief Propagation for Early Vision," International Journal of Computer Vision, Vol. 70, No. 1, pp. 41-54, 2006. https://doi.org/10.1007/s11263-006-7899-4
  8. J. Sun, N. N. Zheng and H. Y. Shum, "Stereo Matching Using Belief Propagation," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 25, No. 7, pp. 203-212, Jul. 2003.
  9. K. Konolige, "Small vision systems: Hardware and implementation," Robotics Research, Springer London, pp. 203-212, 1998.
  10. R. C. Gonzalez, R. E. Woods and S. L. Eddins, Digital Image Processing using MATLAB, McGraw-Hill, 2011.
  11. P. Corke, Robotics, Vision and Control, Springer, 2011.
  12. C. Hwang and F. C. H. Rhee, "Interval Type-2 Fuzzy C Clustering for Detecting Spherical Shells," Journal of Korean Institute of Intelligent Systems, Vol. 14, No. 6, pp. 713-719, 2004. https://doi.org/10.5391/JKIIS.2004.14.6.713
  13. K. P. Murphy, Machine Learning: A Probabilistic Perspective, The MIT Press, 2012.
  14. N. Ho, "Loopy Belief Propagation, Markov Random Field, Stereo Vision", Available: http://nghiaho.com/?page_id=1366, 2012, [Accessed: August 24, 2015]
  15. B. C. Huang and T. Jebara, "Loopy belief propagation for bipartite maximum weight b-matching," Proceedings of International Conference on Artificial Intelligence and Statistics, pp. 195-202, 2007.
  16. A. Klaus, M. Sormann and K. Karner, "Segmented-Based Stereo Matching Using Belief Propagation and a Self-Adapting Dissimilarity Measure," Proceedings of the 18th International Conference on Pattern Recognition, Vol. 3, pp. 15-18, 2006.
  17. Y. Weiss and W. T. Freeman, "On the optimality of solutions of the max-product belief propagation algorithm in arbitrary graphs," IEEE Transactions on Information Theory, Vol. 47, No. 2. pp. 736-744, 2001. https://doi.org/10.1109/18.910585