doi: 10.4304/jsw.9.2.280-286
A Fast Global Minimization of Region-Scalable Fitting Model for Medical Image Segmentation
- 1 College of Information Science and Engineering, Shandong University of Science and Technology, Qingdao, China
- 2 College of Precision Instrument and Opto-Electronics Engineering, Tianjin University, Tianjin, China
Abstract
Active contour model (ACM) which has beenextensively studied recently is one of the most successfulmethods in image segmentation. The present paperadvances an improved hybrid model based on Region-Scalable Fitting Model by combining global convexsegmentation method with edge detector operator. Theproposed model not only inherits the ability of RSF modelto deal with the images with intensity inhomogeneity, butalso overcomes such a drawback: existence of local minimabecause of non-convexity that makes the segmentationresult highly dependent of the initial position of the contour.In addition, the paper exploits two fast numericalimplementation schemes to overcome a huge amount oflevel set methods. The duality projection method isimplemented by introducing dual variables which lead tosemi-implicit iterative scheme of dual variables as well asexact formulation of primal variables. The Split-Bregmanmethod is implemented by introducing auxiliary variableswhich transform the relaxed convex model into solvingsimple poisson equations and exact soft thresholdingformulation. Experimental results for synthetic and realmedical images prove that the proposed model is featuredby greater numerical accuracy and faster division speed.
Keywords
- Region-Scalable fitting model
- global co
How to Cite
Chang-Lei Dongye, Yong-Guo Zheng, and Dong-Huan Jiang, "A Fast Global Minimization of Region-Scalable Fitting Model for Medical Image Segmentation," Journal of Software, vol. 9, no. 2, pp. 280-286, 2014. https://doi.org/ 10.4304/jsw.9.2.280-286
Copyright & License
Copyright © 2014 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).