A differential evolutionary algorithm for multi-threshold image segmentation based on adaptive parameter control strategy DOI Creative Commons

Zong-Na Zhu,

Zhao‐Guang Liu, Ning Wang

и другие.

Research Square (Research Square), Год журнала: 2023, Номер unknown

Опубликована: Окт. 17, 2023

Abstract Multi-threshold image segmentation is a simple and effective approach. Image techniques are significant in the fields of pattern recognition computer vision. However, as number thresholds increases, temporal complexity selecting best threshold increases exponentially. A meta-heuristic optimization approach called differential evolution (DE) algorithm was utilized to address problem. This paper proposes an enhanced DE with adaptive control parameters (IJADE) for multi-threshold segmentation. In this study, optimizes five distinct eight standard test images using maximum between-class variance (OTSU) technique objective function. Comparison analysis IJADE other benchmark algorithms demonstrated viability efficiency proposed method. The quantitative findings demonstrate that peak signal-to-noise ratio structural similarity index measure results under various can be significantly improved by compared existing methods. Peak ratios fabric crane were 22.197 23.1786, respectively, at 5, both placing top. With superior performance digital segmentation, proven more effective.

Язык: Английский

Multi-threshold segmentation of liver tumor images based on deep learning and improved whale optimization algorithm DOI
新一郎 本橋, Xueling Long, Yongqi Li

и другие.

Опубликована: Авг. 13, 2024

Язык: Английский

Процитировано

0
A differential evolutionary algorithm for multi-threshold image segmentation based on adaptive parameter control strategy DOI Creative Commons

Zong-Na Zhu,

Zhao‐Guang Liu, Ning Wang

и другие.

Research Square (Research Square), Год журнала: 2023, Номер unknown

Опубликована: Окт. 17, 2023

Abstract Multi-threshold image segmentation is a simple and effective approach. Image techniques are significant in the fields of pattern recognition computer vision. However, as number thresholds increases, temporal complexity selecting best threshold increases exponentially. A meta-heuristic optimization approach called differential evolution (DE) algorithm was utilized to address problem. This paper proposes an enhanced DE with adaptive control parameters (IJADE) for multi-threshold segmentation. In this study, optimizes five distinct eight standard test images using maximum between-class variance (OTSU) technique objective function. Comparison analysis IJADE other benchmark algorithms demonstrated viability efficiency proposed method. The quantitative findings demonstrate that peak signal-to-noise ratio structural similarity index measure results under various can be significantly improved by compared existing methods. Peak ratios fabric crane were 22.197 23.1786, respectively, at 5, both placing top. With superior performance digital segmentation, proven more effective.

Язык: Английский

Процитировано

0