JACIII Vol.16 No.6 pp. 677-686
doi: 10.20965/jaciii.2012.p0677


Optimal Parameter Setting of Active-Contours Using Differential Evolution and Expert-Segmented Sample Image

Arman Darvish and Shahryar Rahnamayan

Faculty of Engineering and Applied Science, University of Ontario Institute of Technology (UOIT), 2000 Simcoe Street North, Oshawa, Ontario L1H 7K4, Canada

October 9, 2011
June 19, 2012
September 20, 2012
image segmentation, object extraction, active contour (Snake), differential evolution (DE), ultrasound
Generally, tissue extraction (segmentation) is one of the most challenging tasks in medical image processing. Inaccurate segmentation propagates errors to the subsequent steps in the image processing chain. Thus, in any image processing chain, the role of segmentation is in fact critical because it has a significant impact on the accuracy of the final results, such as those of feature extraction. The appearance of variant noise types makes medical image segmentation a more complicated task. Thus far, many approaches for image segmentation have been proposed, including the well-known active contour (snake) model. This method minimizes the energy associated with the target’s contour, which is the sum of the internal and external energy. Although this model has strong characteristics, it suffers from sensitivity to its control parameters. Finding the optimal parameter values is not a trivial task, because the parameters are correlated and problem-dependent. To overcome this problem, this paper proposes a new approach for setting snake’s optimal parameters, which utilizes an expertsegmented gold (ground-truth) image and an optimization algorithm to determine the optimal values for snake’s seven control parameters. The proposed approach was tested on three different medical image test suites: prostate ultrasound (33 images), breast ultrasound (30 images), and lung X-Ray images (48 images). In the current approach, the DE algorithm is employed as a global optimizer. The scheme introduced in this paper is general enough to allow snake to be replaced by any other segmentation algorithm, such as the level set method. For experimental verification, 111 images were utilized. In a comparison with the prepared gold images, the overall error rate is shown to be less than 3%. We explain the proposed approach and the experiments in detail.
Cite this article as:
A. Darvish and S. Rahnamayan, “Optimal Parameter Setting of Active-Contours Using Differential Evolution and Expert-Segmented Sample Image,” J. Adv. Comput. Intell. Intell. Inform., Vol.16 No.6, pp. 677-686, 2012.
Data files:
  1. [1] R. C. Gonzalez and R. E. Woods, “Thresholding. In Digital Image Processing,” pp. 595-611, Pearson Education, 2002. ISBN: 81-7808-629-8
  2. [2] J. C. Dunn, “Well Separated Clusters and Optimal Fuzzy Partitions,” J. Cybern., Vol.4, No.3, pp. 95-104, 1974.
  3. [3] K. Qin, K. Xu, Y. Du, and D. Li, “Fuzzy Systems and Knowledge Discovery (FSKD),” 2010 Seventh Int. Conf. on, Aug. 10-12, 2010, Yantai, Shandong, pp. 524-528, 2010.
  4. [4] J. Canny, “A computational approach to edge detection,” IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol.8, pp. 679-714, 1986.
  5. [5] M. Kass, A. Witkin, and D. Terzopoulos, “Snakes: Active Contour Models,” Int. J. of Computer Vision, Vol.1, pp. 321-332, 1988.
  6. [6] S. Osher and N. Paragios, “Geometric Level Set Methods in Imaging Vision and Graphics,” Springer Verlag, 2003. ISBN: 0387954880
  7. [7] J. Shi and J. Malik, “Normalized Cuts and Image Segmentation,” IEEE Trans. on pattern analysis and machine intelligence, Vol.22, No.8, pp. 888-905, 2000.
  8. [8] M. Khelif, F. Derraz, and M. Beladgham, “Application of Active Contour Models in Medical Image Segmentation,” Int. Conf. on Information Technology: Coding and Computing (ITCC’04), Vol.2, April 2004, Nevada, p. 675, 2004.
  9. [9] L. T. Kejun, W. G. Sheng, and F. Yonghual, “An Image Segmentation Algorithm Based on the Simulated Annealing and Improved Snake Model,” Mechatronic and automation, Int. Conf., Harbin, pp. 3876, August 2007.
  10. [10] J. H. Tan, N. G. Eyk, and R. Acharya, “Detection of Eye and Cornea on IR Thermogram Using Genetic Snake Algorithm,” 9th Int. Conf. on Quantitative InfraRed Thermography, July 2-5, 2008, Krakow, Poland, pp. 21-36, 2008.
  11. [11] M. Gastaud, M. Barlaud, and G. Aubert, “Tracking Video Objects Using Active Contours,” Proc. of Workshop on Motion and Video Computing, pp. 90-95, 2002.
  12. [12] S. Rahnamayan and Z. S. Mohamad, “Breast Ultrasound Segmentation Based on Evolutionary Optimization of Image Processing Chain,” Int.Workshop on Real Time Measurement, Instrumentation and Control [RTMIC], UOIT, Oshawa, Canada, June 25-26, 2010.
  13. [13] S. Rahnamayan, H. R. Tizhoosh, and M. M. A. Salama, “Automatic Acquisition of Image Filtering and Object Extraction Procedures from Ground-Truth Samples,” J. of Advanced Computational Intelligence and Intelligent Informatics, Vol.13, No.2, pp. 115-127, 2009.
  14. [14] C. Liu, J. Ma, and G. Ye, “Medical Image Segmentation by Geodesic Active Contour Incorporating Region Statistical Information,” Fourth Int. Conf. on Fuzzy Systems and Knowledge Discovery, IEEE, 2007.
  15. [15] L. Wang, Ch. Li, Q. Sun, D. Xia, and C.-Y. Kao, “Active Contours Driven by Local and Global Intensity Fitting Energy with Application to Brain MR Image Segmentation,” Elsevier Computerized Medical Imaging and Graphics, pp. 520-531, 2009.
  16. [16] T. McInerney and D. Terzopoulos, “Medical Image Segmentation Using Topologically Adaptable Snake,” Int. Conf. on Computer Vision, Virtual Reality and Robotics in Medicine, Grenoble, France, March, 1997.
  17. [17] R. Boscolo, M. S. Matthew, S. Brown, and M. F. McNitt-Gray, “Medical Image Segmentation with Knowledge-guided Robust Active Contours,” infoRAD on March-April 2002, Vol.22, pp. 437-448, 2002.
  18. [18] D. J.Williams and M. Shah, “A Fast Algorithm for Active Contours and Curvature Estimation,” Department of Computer Science, University of Central Florida, Orlando, Florida 32816, CVGIP; Image Understanding, Vol.55, No.1, pp. 14-26, January 1992.
  19. [19] B. van Ginneken, A. F. Frangi, J. J. Staal, M. M. ter Haar Romeny, and M. A. Viergever, “Active Shape Model Segmentation with Optimal Features,” IEEE Trans. on Medical Imaging, Vol.21, No.8, pp. 924-933, August 2002.
  20. [20] J.-J. Rousselle, N. Vincent, and N. Verbeke, “Genetic Algorithm to Set Active Contour,” 10th Int. Conf. Computer Analysis of Images and Patterns CAIP-2003, Tours, France, August 25-27, Groningen, Hollande, pp. 345-352, 2003.
  21. [21] K. V. Price, R. M. Storn, and J. A. Lampinen, “Differential Evolution: A Practical Approach to Global Optimization,” Springer, 2005. ISBN: 3540209506

*This site is desgined based on HTML5 and CSS3 for modern browsers, e.g. Chrome, Firefox, Safari, Edge, Opera.

Last updated on Jul. 23, 2024