Related Experiment Video
Updated: Jan 16, 2026
04:48
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
3.3K
ACPOA: An Adaptive Cooperative Pelican Optimization Algorithm for Global Optimization and Multilevel Thresholding
YuLong Zhang1, Jianfeng Wang1, Xiaoyan Zhang1
1South Korea College of Design, Hanyang University, Ansan 15588, Republic of Korea.
Biomimetics (Basel, Switzerland)
|September 26, 2025
Summary
This study introduces the Adaptive Cooperative Pelican Optimization Algorithm (ACPOA) to improve image segmentation. ACPOA enhances accuracy and efficiency in complex tasks like medical imaging, offering a superior solution.
Area of Science:
- Computer Vision
- Artificial Intelligence
- Optimization Algorithms
Background:
- Multi-threshold image segmentation is crucial for accurate target detection and regional analysis in fields like medical imaging and remote sensing.
- Existing segmentation algorithms often exhibit slow convergence and low accuracy, limiting their practical application.
- Developing efficient and accurate optimization algorithms is essential for advancing image segmentation.
Purpose of the Study:
- To propose and evaluate the Adaptive Cooperative Pelican Optimization Algorithm (ACPOA) for global optimization and multilevel threshold image segmentation.
- To address the limitations of existing algorithms, specifically slow convergence and low solution accuracy.
- To enhance the performance of image segmentation in complex real-world applications.
Main Methods:
- Developed ACPOA by integrating an elite pool mutation strategy, an adaptive cooperative mechanism for high-dimensional search, and a hybrid boundary handling technique.
- Tested ACPOA against eight advanced algorithms on the CEC2017 and CEC2022 benchmark test suites for global optimization.
- Applied ACPOA to multilevel threshold image segmentation tasks to assess its practical performance.
Main Results:
- ACPOA demonstrated superior optimization performance compared to eight advanced algorithms on benchmark test suites.
- Experiments validated the effectiveness of the elite pool mutation, adaptive cooperative mechanism, and hybrid boundary handling strategies.
- ACPOA achieved better accuracy, stability, and efficiency in multilevel threshold image segmentation tasks.
Conclusions:
- The proposed ACPOA offers significant improvements in both global optimization and multilevel threshold image segmentation.
- ACPOA provides an effective solution for complex optimization challenges in image analysis.
- The algorithm's enhanced accuracy and efficiency make it suitable for critical applications in medical imaging, remote sensing, and industrial inspection.
