Nature-inspired optimization algorithms and their significance in multi-thresholding image segmentation: an inclusive...

Area of Science:

• Computer Vision and Image Processing
• Artificial Intelligence
• Optimization Techniques

Background:

• Multilevel Thresholding (MLT) is crucial for segmenting complex images with nonlinear conditions.
• MLT is an optimization problem often requiring nondeterministic solutions.
• Nature-Inspired Optimization Algorithms (NIOA) are increasingly used for MLT.

Purpose of the Study:

• To provide a review of novel Nature-Inspired Optimization Algorithms (NIOA) developed between 2019-2021 for Multilevel Thresholding (MLT).
• To highlight and explore the challenges in developing MLT models using NIOA.

Main Methods:

• Literature review of recent (2019-2021) Nature-Inspired Optimization Algorithms.
• Analysis of the application of these algorithms to Multilevel Thresholding problems.
• Identification of common challenges in NIOA-based MLT model development.

Main Results:

• Several novel NIOAs have emerged in the last three years.
• These algorithms offer potential solutions for complex image segmentation tasks.
• Specific challenges related to the application of NIOA in MLT have been identified.

Conclusions:

• NIOAs are a promising approach for addressing the complexities of Multilevel Thresholding.
• Further research is needed to overcome identified challenges in developing robust NIOA-based MLT models.
• The review provides insights into recent advancements and future directions in the field.