结合非局部空间信息和KL信息的鲁棒FCM算法
A Robust FCM Algorithm Combining Non-Local Spatial Information and KL Information
彭家磊 1黄成泉 2陈阳 1雷欢 1覃小素1
作者信息
- 1. 贵州民族大学数据科学与信息工程学院,贵州 贵阳 550025
- 2. 贵州民族大学工程技术人才实践训练中心,贵州贵阳 550025
- 折叠
摘要
针对传统模糊C均值(Fuzzy C-Means,FCM)聚类算法对噪声敏感的问题,提出一种结合非局部空间信息和KL信息的鲁棒FCM算法.首先,将灰度信息与非局部空间信息相融合,用于增强算法对噪声的鲁棒性;其次,在目标函数中引入KL信息,以便减少分割的模糊性.在密度为5％的混合噪声条件下,合成图像和自然图像的实验结果表明,该文算法的分割精度较高、鲁棒性较强,能较好地分割噪声图像.
Abstract
Aiming at the problem that traditional Fuzzy C-Means(FCM)clustering algorithm is sensitive to noise,a robust FCM algorithm combining non-local spatial information and KL information is proposed.Firstly,the gray-scale information is fused with non-local spatial information to enhance the robustness of the algorithm against noise.Secondly,KL information is introduced into the objective function to reduce the ambiguity of segmentation.Under the mixed noise condition of 5％density,the experimental results of synthesized image and natural image show that the algorithm has high segmentation accuracy and robustness,and can divide noise image better.
关键词
模糊C均值/图像分割/非局部空间信息/KL信息Key words
Fuzzy C-Means/Image segmentation/Non-local spatial information/KL information引用本文复制引用
出版年
2024
NETL
NSTL
万方数据