Localised active contour method via local similarity measure for image segmentation
by Xiaoliang Jiang; Jinyun Jiang
International Journal of Data Science (IJDS), Vol. 7, No. 3, 2022

Abstract: The accuracy of active contour methods is not always exact since there are many uncertainty factors, e.g., abundant noise, lack of clear boundaries, intensity inhomogeneity. To tackle these issues, a localised region-based segmentation framework is presented in this paper. In our method, a new adaptive local similarity measure is built in local regions as the spatial constraint to guarantee noise suppression and outlier resistance. Second, we construct an objective equation by integrating the local similarity measure into an active contour algorithm based on the local region. Furthermore, we design the local mean difference energy as a control constraint to enhance the efficiency and smoothness of the profile curve. Experimental data demonstrate that our algorithm, when compared with other classical region-based models, can achieve higher accuracy and has stronger robustness for images with higher noise levels.

Online publication date: Wed, 14-Dec-2022

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Data Science (IJDS):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com