Abstract
A computational method is presented which efficiently segments digital grayscale images by directly applying the Q-state Ising (or Potts) model. Since the Potts model was first proposed in 1952, physicists have studied lattice models to gain deep insights into magnetism and other disordered systems. For some time, researchers have realized that digital images may be modeled in much the same way as these physical systems (i.e., as a square lattice of numerical values). A major drawback in using Potts model methods for image segmentation is that, with conventional methods, it processes in exponential time. Advances have been made via certain approximations to reduce the segmentation process to power-law time. However, in many applications (such as for sonar imagery), real-time processing requires much greater efficiency. This article contains a description of an energy minimization technique that applies four Potts (Q-Ising) models directly to the image and processes in linear time. The result is analogous to partitioning the system into regions of four classes of magnetism. This direct Potts segmentation technique is demonstrated on photographic, medical, and acoustic images.
Similar content being viewed by others
References
D. L. Pham, C. Xu, J. L. Prince, Annu. Rev. Biomed. Eng. 2, 315 (2000)
Y. H. Yang, M. J. Buckley, S. Dudoit, T. P. Speed, J. Comput. Graph. Stat. 11, 108 (2002)
S. Peng, B. Urbanc, L. Cruz, B. T. Hyman, H. E. Stanley, P. Natl. Acad. Sci. USA 100, 3847 (2003)
V. Grau, A. U. J. Mewes, M. Alcañiz, IEEE T. Med. Imaging 23, 447 (2004)
S. Hadjidemetriou, C. Studholme, S. Mueller, M. Weiner, N. Schuff, Med. Image Anal. 13, 36–48 (2009)
X. Descombes, M. Moctezuma, H. Maître, J.-P. Rudant, Signal Process. 55, 123–132 (1996)
F. W. Bentrem, W. E. Avera, J. Sample, Sea Technol. 47, 37 (2006)
T. Asano, D. Z. Chen, N. Katoh, T. Tokuyama, Int. J. Comput. Geom. Ap. 11, 145 (2001)
E. Ising, Z. Phys. 31, 253 (1925)
R. B. Potts, P. Camb. Philos. Soc. 48, 106 (1952)
F. W. Bentrem, Provisional Patent Application, Navy Case No. 99, 755 (2009)
K. Tanaka, J. Phys. A-Math. Gen. 35, R81 (2002)
J. P. Neirotti, S. M. Kurcbart, N. Caticha, Phys. Rev. E 68, 031911 (2003)
M. Blatt, S. Wiseman, E. Domany, Phys. Rev. Lett. 76, 3251 (1996)
M. Blatt, S. Wiseman, E. Domany, Neural Computation 9, 1805 (1997)
S. Wiseman, M. Blatt, E. Domany, Phys. Rev. E 57, 3767 (1998)
K. Tanaka, H. Shouno, M. Okadak, D. M. Titterington, J. Phys. A-Math. Gen. 37, 8675 (2004)
E. Sharon, A. Brandt, R. Basriy, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (IEEE, Hilton Head Island, 2000) 1, 70
P. F. Felzenszwalb, D. P. Huttenlocher, Int. J. Comput. Vision 59, 167 (2004)
A. X. Falcão, P. A. V. Miranda, A. Rocha, Lect. Notes Comp. Sci. 4179/2006, 138 (2006)
D. Chandler, Introduction to Modern Statistical Mechanics (Oxford University Press, New York, 1987)
J. C. Lee Thermal Physics: Entropy and Free Energies (World Scientific Publishing Company, Singapore, 2002)
L. Onsager, Phys. Rev. 65, 117 (1944)
D. Martin, C. Fowlkes, D. Tal, J. Malik, Proceedings of the 8th International Conference on Computer Vision 2, 416 (2001)
F. W. Bentrem, J. Sample, M. T. Kalcic, M. E. Duncan, Proceedings of Oceans 2002 (MTS/IEEE, Biloxi) 1, 7 (2002)
F. W. Bentrem, J. T. Sample, M. M. Harris, Scientific Computing 25, 30 (2008)
R. A. Bagnold, The Physics of Blown Sand and Desert Dunes (Methuen, London, 1941)
L. Kang, L. Guo, Phys. Lett. A 330, 198 (2004)
Author information
Authors and Affiliations
Corresponding author
About this article
Cite this article
Bentrem, F.W. A Q-Ising model application for linear-time image segmentation. centr.eur.j.phys. 8, 689–698 (2010). https://doi.org/10.2478/s11534-009-0165-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.2478/s11534-009-0165-y