Fuzzy Clustering Algorithm Based on Improved Global Best-Guided Artificial Bee Colony with New Search Probability...

Area of Science:

• Computer Vision
• Artificial Intelligence
• Image Processing

Background:

• Fuzzy C-means (FCM) is a widely used soft segmentation technique for image analysis.
• FCM's effectiveness is hindered by sensitivity to initial cluster centers, noise, and a tendency to converge to local optima.

Purpose of the Study:

• To address the limitations of traditional FCM clustering algorithms.
• To enhance the exploration and exploitation balance in FCM for improved image segmentation.

Main Methods:

• A two-phase approach integrating an improved global best-guided artificial bee colony algorithm (IABC) with a new search probability model (PIABC).
• The PIABC algorithm balances exploration and exploitation to refine FCM's initial cluster center selection.
• The proposed PIABC-FCM method utilizes PIABC's balancing capabilities to avoid local optima during FCM clustering.

Main Results:

• The PIABC-FCM algorithm demonstrated improved performance in grayscale image segmentation.
• The proposed method showed promising results compared to existing related works, indicating enhanced accuracy and robustness.

Conclusions:

• The PIABC-FCM approach effectively mitigates the local optima and noise sensitivity issues inherent in FCM.
• This novel method offers a robust and accurate solution for grayscale image segmentation, outperforming conventional techniques.