4.6 Article

Chaotic multi verse improved Harris hawks optimization (CMV-IHHO) facilitated multiple level set model with an ideal energy active contour for an effective medical image segmentation

Journal

MULTIMEDIA TOOLS AND APPLICATIONS
Volume 81, Issue 15, Pages 20963-20992

Publisher

SPRINGER
DOI: 10.1007/s11042-022-12344-x

Keywords

Contour; Chan and Vese; Level set; Meta-heuristic optimization; Multi verse; Harris hawks optimization

Ask authors/readers for more resources

This manuscript proposes a multi-level set model with an ideal energy active contour for image segmentation. The proposed approach addresses the issue of inappropriate initialization and achieves acceptable segmentation performance. The method utilizes meta-heuristic optimization algorithm and employs Chaotic Multi Verse Improved Harris Hawks Optimization (CMV-IHHO) technique to solve the energy minimization problem.
Nowadays, the contour models (CMs) are widely used in image segmentation. Among these CMs, the Chan and Vese model depending on level set is the current regional based model, considering the regularity of intensity in every region. If the contour is not initialized correctly, the conventional level set (LS) model frequently sticks in local minima. This is becoming more important in the medical images context. In this manuscript, a multi-level set model with an ideal energy active contour is proposed, which is anticipated to realize the performance of acceptable segmentation, regardless of the contour's initial choice. The active contour models are utilized to identify an object outline from the image. The active CMs with energy based segmentation methods minimizes the energy related with active contour. This work makes the appropriate energy minimized difficult to solve by using meta-heuristic optimization algorithm and makes a proficient execution of the approach by Chaotic Multi Verse Improved Harris Hawks Optimization (CMV-IHHO) technique. Here, the proposed approach is compared with six existing approaches. The existing methods such as DA, Symmetry Analysis, Fuzzy C-Means, Rough Fuzzy C-Means, K-Means Level Set, Random Forest, and Support vector machine method. The accuracy of the proposed method is 0.91%, 5.84%, 15.63%, 8.30%, 10.97%, 15.77%, and 5.14% better than the existing approaches. The sensitivity of the proposed method is 3.37%, 5.74%, 22.66%, 4.54%, 17.94%, 4.54% and 15% better than the existing methods.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.6
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available