A Novel GVF-Based Adaptive Extracting Formulation for Image Reconstruction

Abstract:

This paper proposed a fast boundary reconstruction algorithm based on adaptive gradient vector flow (GVF) deformable model. The novel dynamic GVF field performs faster and wider capture range over original GVF model, which efficiently attracts the model approaching the object contour more quickly and more stably. Based on the approximation error analysis the algorithm can automatically add new knots in contour curve, which makes the accelerated model achieve adaptive adjustment according to local characteristics of the boundary. The improved algorithm can greatly increase the reconstruction accuracy without compromising approximation efficiency. Additional experiments demonstrate the efficient procedure and fine performance of the approach.

• Periodical: Applied Mechanics and Materials (Volumes 239-240)
• Pages: 1250-1254
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.239-240.1250
• Online since: December 2012
• Authors: Bin Zhao, Qin Liang Li, Ying Chun Wang, Guo Xing Jiang
• Keywords: Boundary Extraction, B-Spline, Finite Element Method (FEM), GVF Snake

References

[1] M. Kass, A.T. Witkin, D. Terzopoulos. Snakes, Active contour models. International Journal of computer vision, vol. 1(4), 1988, 321-331.

DOI: 10.1007/bf00133570

[2] J. Jeong, K. Kim, H. Park, and et al, B-Spline surface approximation to cross-sections using distance maps, International journal, advanced manufacturing technology , vol. 15, 1999, 876-885.

DOI: 10.1007/s001700050145

[3] G. Zhu, S. Zhang, Q. Zeng, and et al. Gradient vector flow active contours with prior directional information. Pattern Recognition Letters, 2010, 31. 845-856.

DOI: 10.1016/j.patrec.2010.01.011

[4] Ghita, P.F. Whelan. A new GVF-based image enhancement foumulation for use in the presence of mixed noise. Pattern Recognition, 2010, 43. 2646-2658.

DOI: 10.1016/j.patcog.2010.02.023

[5] L.D. Cohen. On Active Contour Models and Balloons. CVGIP: Image Understanding, vol 53(2): 211-218, (1991).

DOI: 10.1016/1049-9660(91)90028-n

[6] H. Li, Y. Ching, D.B. Tien, and P. Zhang. Realization of handwritten numerical recognition based on constant curvature character, TUXIANGSHIBIE YU ZIDONGHUA, vol. 2(3), 2005, 152-154.

[7] C. Xu and J.L. Prince. Snakes, shapes, and gradient vector flow, IEEE Trans. On Image Processing, vol. 7, 1998, pp.359-369.

DOI: 10.1109/83.661186

[8] C. Xu, L. Jerry. Generalized gradient vector flow external forces for active contours, Signal Processing, vol. 77, 1998, 131-139.

DOI: 10.1016/s0165-1684(98)00140-6

[9] S. Luo. Automated Medical Image Segmentation Using a New Deformable Surface Model. International Journal of Computer Science and Network Security, vol. 6(5), 2006, 109-115.

[10] S. Cheng, X. Zhang, C. Jin. Finite element method based B-spline active contour. Journal of information and computational science, 2004, 1(2): 275-280.