قطعه‌بندی تصویر مبتنی بر برش نرمالیزه گراف از دیدگاه میزان اطلاعات جداکننده

نویسندگان

1 دانشگاه صنعتی نوشیروانی بابل - دانشکده مهندسی برق و کامپیوتر

2 دانشگاه شهید بهشتی - دانشکده مهندسی هسته‌ای

چکیده

قطعه‌بندی تصویر، یک مسئله پایه در بینایی ماشین است. در روش مبتنی بر برش نرمالیزه گراف (Ncut)، حل این مسئله به انتخاب بردار ویژه متناظر با دومین کوچک‌ترین مقدار ویژه یک ماتریس خاص می‌انجامد. در این مقاله، ضمن بیان هم‌ارزی رابطه ریاضی حاکم بر مسئله بدون مربیِ Ncut با معیار Fisher-Rao در طبقه‌بندیِ با مربی، از نگاهی نو به مسئله انتخاب بردار ویژه پرداخته شده است. در این مقاله با پیشنهاد معیاری کـارا از دیدگاه Fisher-Rao، گزینش و مرتب‌سازی بردارهای ویژه در مسئله هم‌ارزِ Ncut آن انجام شده است. نتایج آزمایش هم‌ارزی قطعه‌بندی تصویر بر پایه این دو معیار، ارائه قطعه‌بندی با اندازه Ncut کمتر و گوناگونیِ ارزش‌گذاریِ بردارهای ویژه را نشان می‌دهد.

کلیدواژه‌ها


[1] B. Peng, L. Zhang, and  D. Zhang, “A survey of graph theoretical approach to image segmentation,” Paternn Recognition, 2012.
[2] R.C. Gonzalez, and R.E. Wood, Digital Image Processing,Prentice Hall, New York, USA, 2002.
[3] S. Naz, H. Majee, and H. Irshad, “Image segmentation using fuzzy clustering: A survey,” International Conference on Emerging Technologies, pp. 181-186, 2010.
[4] S. Raut, M. Raghuvanshi, and R. Dharaskar, A. Raut, “Image segmentation: a state-of-art survey for prediction,” ICACC, pp. 420-424, 2009.
[5] L. Wang, “Comparison for edge detection of colony image,” IJCSNS International Journal of Computer Science and Network Security, vol. 6 no. 9A, 2006.
[6] Z. Musoromy, R. Soodamani, and B. Nico, “Edge detection comparison for License plate detection,” Int. Control, Automation, Robotics and Vision, 2010.
[7] I. Karoui, J. Boucher, and J. Augustin, “Variational region-Based segmentation using multiple texture statistics,” IEEE Trans on Image Processing, vol. 19, no. 12, 2010.
[8] C. Rambabu, I. Chakrabarti, and A. Mahanta, “Flooding-based watershed algorithm and its prototype hardware architecture,” IEEE Proceedings Vision Image and Signal Processing, vol. 151, no. 3, pp. 224-234, 2004.
[9] A.X. Falcao, J.K. Udupa, and F.K. Miyazawa, “An ultra-fast user-steered image segmentation paradigm: live wire on the fly,” IEEE Trans on Medical Imaging, vol. 19, no. 1, pp. 55-62, 2002.
[10] G. Sundaramoorthi, A. Yezzi, and A.C. Mennucci, “Coarse-to-Fine segmentation and tracking using sobolev active contours,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 30, no. 5, pp. 851-864, 2008.
[11] R.M. Willett, and R.D. Nowak, “Minimax optimal level-set estimation,” IEEE Trans. on Image Processing, vol. 16, no. 12, pp. 2965-2979, 2007.
[12] H. Yu, D. Wang, and Z. Tan, “Level set methods and image segmentation,” International Workshop on Medical Imaging and Augmented Reality, pp. 204-208, 2002.
[13] Y.B. Yuri, and G.F. Lea, “Graph cuts and efficient N-D image segmentation,” International Journal of Computer Vision, vol. 70, no. 2, pp. 109-131, 2006.
[14] Y.B. Yuri, and M.P. Jolly, “Interactive graph cuts for optimal boundary & region segmentation of objects in N-D images,” Proceedings of Internation Conference on Computer Vision, vol. I, pp. 105-112, 2001.
[15]  J.B. Kruskal. “On the shortest spanning subtree of a graph and the traveling salesman problem,” Proceedings of the American Mathematical Society, vol. 7, no. 1, pp. 48-50, 1956.
[16] Z. Wu, and R. Leahy, “Tissue classification in MR images using hierarchical segmentation,” Proc. IEEE Int. Conf: Medical Imaging, vol. 12, no. 1, pp. 81-85, 1990.
[17] D. German, and B Gidas, Image Analysis and Computer Vision, National Academy Press, Chapter 2, pp.9-36, 1991.
[18] E.W. Dikstra, “Some theorems on spanning subtrees of a graph,” Indag Math., vol. 22, no. 2, pp. 196-199, 1960.
[19] L.Grady. “Multilabel random walker segmentation using prior models,” IEEE Conference of Computer Vision and Pattern Recognition, vol. 1, pp. 763-770, 2005.
[20] J. Shi, and J. Malik, “Normalized cut and image segmentation,” IEEE Transaction On Pattern Analysis and Machine Inteligence, vol. 22, no. 8, 2000.
[21]S.D. Inderjit, Y. Guan, and B. Kulis, “Weighted graph cuts without eigenvectors: a multilevel approach,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 29, no. 11, 2007.
[22] S. Wang, and J.M. Siskind; “Image segmentation with ratio cut,” IEEE Transaction On Pattern Analysis and Machine Intelligence, vol. 25; no. 6, 2003.
[23]E.R. Pujadas, and M. Reisert, “Shape-based normalized cuts using spectral relaxation for biomedical segmentation,” Image Processing, vol. 23, no. 1, pp. 163-170, 2014.
[25] Z.Y. Wu, and R. Leahy, “An optimal graph theoretic approach to data clustering: theory and its application to image segmentation,” IEEE Trans. Pattern Anal. Mach.Intell., vol. 15, no. 11, pp. 1101-1113, 1993.
[28] D. Wangnerand, and F. Wangner, Between Min Cut and Graph Bisection, TU Berlin, Brlin, Germany, Algorithmic Discrete Mathematics, Technical Report, 1991.
[29] A. Pothen, and H. Simon, “Partitioning sparse matrices with eigenvectors of graph,” IBM Journal of Research and Development, pp. 420-425, 1973.
[30]  M. Zhu, and A.M. Martinez, “Selecting principal components in a two-stage LDA algorithm,” IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2006.
[31] http://www.cis.upenn.edu/~jshi/software/.