可靠性模糊局部信息C均值聚类算法的苗族服饰图像分割  

Images segmentation of Miao clothing based on dependable fuzzy local information C-means clustering algorithm

作　　者：覃小素 黄成泉 雷欢 陈阳 彭家磊 周丽华 QIN Xiaosu;HUANG Chengquan;LEI Huan;CHEN Yang;PENG Jialei;ZHOU Lihua(College of Data Science and Information Engineering,Guizhou Minzu University,Guiyang 550025,China;Engineering Training Center,Guizhou Minzu University,Guiyang 550025,China)

机构地区：[1]贵州民族大学数据科学与信息工程学院,贵阳550025 [2]贵州民族大学工程技术人才实践训练中心,贵阳550025

出　　处：《智能计算机与应用》2023年第10期142-146,共5页Intelligent Computer and Applications

基　　金：国家自然科学基金(62062024);贵州省省级科技计划项目(黔科合基础-ZK[2021]一般342);贵州省研究生教育教学改革重点项目(黔教合YJSJGKT[2021]018);贵州省教育厅自然科学研究项目(黔教技[2022]015)。

摘　　要：传统的模糊C均值(FCM)算法容易受到噪声的影响,难以对具有褶皱、污渍和色彩差异大的苗族服饰图像进行有效的分割。本文提出可靠性模糊局部信息C均值聚类算法,在FCM的目标函数中增加一个模糊因子来度量局部的相似性,充分结合空间信息与灰度信息,提高对噪声的鲁棒性。此外,引入一个模糊不确定性聚类模型,对像素进行可靠性分析,进一步降低噪声点和边缘点的影响。利用含噪的苗族服饰图像进行实验,实验结果表明本文所提算法对含噪的苗族服饰图像分割效果好,对于褶皱、污渍及色彩差异大的苗族服饰图像,本文所提算法都获得最高的划分系数和最低的划分熵,分割质量均优于对比算法。The traditional fuzzy C-means(FCM) algorithm is easy to be affected by noise,and it is difficult to segment the Miao clothing images with folds,stains and color differences effectively.In this paper,the dependable fuzzy local information C-means clustering algorithm is proposed,which adds a fuzzy factor to the objective function of FCM to measure local similarity,fully combines spatial information and gray information,and improves the robustness to noise.In addition,a fuzzy uncertainty clustering model is introduced to analyze the dependability of pixels to further reduce the influence of noise points and edge points.The experimental results show that the proposed algorithm has good segmentation effect on the noisy Miao clothing images.For the Miao clothing images with large differences in folds,stains and colors,the proposed algorithm has the highest partitioning coefficient and the lowest partitioning entropy,and the segmentation quality is better than the comparison algorithm.

关 键 词：模糊C均值 苗族服饰图像 局部信息 不确定性聚类 可靠性 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]