多融合策略改进灰狼优化算法的图像分割

Gray Wolf Optimization Algorithm for Image Segmentation Improved by Multiple Fusion Strategies

摘要: 本文提出了一种多策略融合改进的灰狼优化算法的图像分割。首先，针对灰狼优化算法的收敛速度慢，易陷入局部最优解等问题分别采用Tent混沌映射优化、收敛因子非线性调整、透镜成像反向学习、修改位置更新方程策略来提升了灰狼算法的搜索性能和收敛速度，并通过不同的基准测试函数验证了改进算法的优越性。其次，通过改进的灰狼优化算法获取K均值聚类算法的初始聚类中心，并应用于图像分割当中从而进一步提高图像分割效率。最后，将所提出的方法应用于医学图像进行实验。实验结果表明，与传统的图像分割算法和其他优化算法相比，本文提出的方法明显提高了图像分割的质量和效果，且具有更好的鲁棒性和稳定性。

Abstract: This paper presents a multi-strategy integrated enhancement of the gray wolf optimization algorithm for image segmentation. Initially, various strategies such as Tent chaotic mapping optimization, non-linear adjustment of convergence factor, reverse-learning based lens imaging, and position updating are employed to enhance the search performance and convergence speed of the gray wolf algorithm, addressing issues like slow convergence speed and susceptibility to local optima. The enhanced algorithm is validated through different benchmark test functions. Subsequently, the improved gray wolf optimization algorithm is utilized to obtain initial clustering centers for image segmentation, further enhancing the efficiency of the segmentation process. Finally, the proposed method is applied to medical images in experimental settings. The experimental results demonstrate that compared to traditional image segmentation algorithms and other optimization algorithms, the proposed method significantly improves the quality and effectiveness of image segmentation, exhibiting better robustness and stability.