Scholarlink

<p>Meta-heuristic algorithms represent a widely applied general framework for solving complex optimization problems. However, existing algorithms often face challenges such as slow convergence, parameter sensitivity, and susceptibility to local optima when addressing high-dimensional, nonlinear, and multimodal optimization problems. To address these issues, this paper proposes a meta-heuristic algorithm inspired by sloth behavior—the Sloth Optimization Algorithm (SOA). By integrating energy state decision-making with velocity updates based on neighborhood mechanisms, SOA effectively avoids getting stuck in local optima. Experimental results on the CEC 2017 and CEC 2022 test function sets demonstrate that SOA achieves dominance rates of 79.3% and 66.7%, respectively, significantly outperforming comparison algorithms. Furthermore, SOA delivers remarkable results in multiple engineering optimization problems and remote sensing image segmentation tasks, validating its practicality and efficiency in real-world applications.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

SOA: a sloth optimization algorithm inspired by natural phenomena and its application to image segmentation

  • 期刊论文

  • 发表日期：

    2026年6月17日

查看全文
  • Hanyu Li,
  • Xiaoliang Zhu,
  • Mengkun Li,
  • Ziwei Yang,
  • Xiaofeng Wang,
  • Wenxing Bao

摘要

Meta-heuristic algorithms represent a widely applied general framework for solving complex optimization problems. However, existing algorithms often face challenges such as slow convergence, parameter sensitivity, and susceptibility to local optima when addressing high-dimensional, nonlinear, and multimodal optimization problems. To address these issues, this paper proposes a meta-heuristic algorithm inspired by sloth behavior—the Sloth Optimization Algorithm (SOA). By integrating energy state decision-making with velocity updates based on neighborhood mechanisms, SOA effectively avoids getting stuck in local optima. Experimental results on the CEC 2017 and CEC 2022 test function sets demonstrate that SOA achieves dominance rates of 79.3% and 66.7%, respectively, significantly outperforming comparison algorithms. Furthermore, SOA delivers remarkable results in multiple engineering optimization problems and remote sensing image segmentation tasks, validating its practicality and efficiency in real-world applications.