Image segmentation algorithm based on the Allen-Cahn energy function.

Area of Science:

• Computer Vision
• Image Processing
• Computational Mathematics

Background:

• Image segmentation is crucial for analyzing visual data.
• Existing methods face challenges in efficiency and robustness.

Purpose of the Study:

• Introduce a novel image segmentation algorithm based on the Allen-Cahn energy function.
• Enhance segmentation efficiency and accuracy.
• Develop an optimized algorithm with autonomous parameter selection.

Main Methods:

• Utilized the Allen-Cahn (AC) energy function for local energy feature calculation.
• Employed a sliding window technique to generate an energy matrix.
• Constructed constraints using extreme values in the energy matrix.
• Implemented an optimized version for autonomous parameter selection.

Main Results:

• Demonstrated effectiveness on simple and complex images, with excellent performance in agronomy and medicine.
• Achieved significant time-saving advantages over state-of-the-art methods.
• Validated on classical datasets, showing high accuracy (>95%) and precision, recall, F1 scores (>90%) on the CO-SKEL dataset.
• Maintained high segmentation accuracy (>94%) under noise interference, proving robustness.

Conclusions:

• The proposed Allen-Cahn based algorithm provides an efficient and robust solution for image segmentation.
• The optimized version with autonomous parameter selection enhances applicability and performance.
• The method shows significant potential for real-world applications in various fields.