Convex Image Segmentation Model Based on Local and Global Intensity Fitting Energy and Split Bregman Method
Joint Authors
Source
Journal of Applied Mathematics
Issue
Vol. 2012, Issue 2012 (31 Dec. 2012), pp.1-16, 16 p.
Publisher
Hindawi Publishing Corporation
Publication Date
2012-02-29
Country of Publication
Egypt
No. of Pages
16
Main Subjects
Abstract EN
We propose a convex image segmentation model in a variational level set formulation.
Both the local information and the global information are taken into consideration to get better segmentation results.
We first propose a globally convex energy functional to combine the local and global intensity fitting terms.
The proposed energy functional is then modified by adding an edge detector to force the active contour to the boundary more easily.
We then apply the split Bregman method to minimize the proposed energy functional efficiently.
By using a weight function that varies with location of the image, the proposed model can balance the weights between the local and global fitting terms dynamically.
We have applied the proposed model to synthetic and real images with desirable results.
Comparison with other models also demonstrates the accuracy and superiority of the proposed model.
American Psychological Association (APA)
Yang, Yunyun& Wu, Boying. 2012. Convex Image Segmentation Model Based on Local and Global Intensity Fitting Energy and Split Bregman Method. Journal of Applied Mathematics،Vol. 2012, no. 2012, pp.1-16.
https://search.emarefa.net/detail/BIM-993563
Modern Language Association (MLA)
Yang, Yunyun& Wu, Boying. Convex Image Segmentation Model Based on Local and Global Intensity Fitting Energy and Split Bregman Method. Journal of Applied Mathematics No. 2012 (2012), pp.1-16.
https://search.emarefa.net/detail/BIM-993563
American Medical Association (AMA)
Yang, Yunyun& Wu, Boying. Convex Image Segmentation Model Based on Local and Global Intensity Fitting Energy and Split Bregman Method. Journal of Applied Mathematics. 2012. Vol. 2012, no. 2012, pp.1-16.
https://search.emarefa.net/detail/BIM-993563
Data Type
Journal Articles
Language
English
Notes
Includes bibliographical references
Record ID
BIM-993563