基于改进非洲秃鹫优化算法的脑MRI图像分割  

Brain MRI image segmentation based on improved African vultureoptimization algorithm

作　　者：王豪 凌基伟 陈昊 黄志勇 王岫鑫 WANG Hao;LING Jiwei;CHEN Hao;HUANG Zhiyong;WANG Xiuxin(School of Life Health Information Science and Engineering,Chongqing University of Posts and Telecommunications,Chongqing 400065,P.R.China;Chongqing NO.39 Middle School,Chongqing 400064,P.R.China)

机构地区：[1]重庆邮电大学生命健康信息科学与工程学院,重庆400065 [2]重庆市第三十九中学,重庆400064

出　　处：《重庆邮电大学学报（自然科学版）》2024年第4期687-696,共10页Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition)

基　　金：国家自然科学基金项目(61605021);重庆市教委科学技术研究项目(KJZD-K202300614);中国博士后科学基金第74批面上资助项目(2023MD744138);重庆市自然科学基金面上项目(CSTB2023NSCQ-MSX0839)。

摘　　要：针对非洲秃鹫优化算法(African vulture optimization algorithm,AVOA)在开发能力与探索能力之间存在的失衡问题,提出了一种多策略改进的非洲秃鹫优化算法(improved African vulture optimization algorithm,IAVOA)。采用佳点集初始化种群以增强多样性,引入混合对立学习以强化开发与探索,实施自适应信任度策略以动态调整搜索过程,应用高斯变异来进一步平衡算法的开发和探索能力。仿真结果显示,在12个典型测试函数上,IAVOA相较于对比算法,在收敛速度、求解精度和稳定性方面均显著提升。提出了IAVOA-FCM算法用于小样本数据集的脑磁共振成像(magnetic resonance imaging,MRI)图像分割,通过IAVOA算法强大的全局寻优能力对FCM算法进行优化。在脑MRI图像分割实验中,与其他5种先进的结合算法相比,IAVOA-FCM在分割精度、稳定性等方面均表现出显著优势。Addressing the imbalance between exploitation and exploration capabilities in the African vulture optimization algorithm(AVOA),we propose an improved African vulture optimization algorithm(IAVOA)incorporating multiple strategies.This algorithm employs the good point set method to initialize the population for enhanced diversity,introduces mixed opposition-based learning to strengthen exploitation and exploration,implements an adaptive trustworthiness strategy for dynamically adjusting the search process,and applies Gaussian mutation to further balance the exploitation and exploration capabilities of the algorithm.Simulation results demonstrate that,compared to competing algorithms,IAVOA significantly improves convergence speed,solution accuracy,and stability on 12 typical test functions.Furthermore,this paper proposes the IAVOA-FCM algorithm for brain MRI image segmentation in small sample datasets,optimizing the FCM algorithm through the powerful global optimization capability of the IAVOA.In the experiments for brain magnetic resonance imaging(MRI)image segmentation,compared to five other advanced hybrid algorithms,IAVOA-FCM exhibits significant advantages in segmentation accuracy,stability,and other aspects.

关 键 词：非洲秃鹫优化算法 模糊C均值 群智能优化算法 信任度策略 脑磁共振成像(MRI)图像分割 

分 类 号：TN011[电子电信—物理电子学]