[1]温金玉,宣士斌,黄亚武,等.多尺度多类中性模糊聚类图像分割算法[J].计算机技术与发展,2019,29(07):65-70.[doi:10. 3969 / j. issn. 1673-629X. 2019. 07. 013]
　WEN Jin-yu,XUAN Shi-bin,HUANG Ya-wu,et al.Image Segmentation Algorithm Based on Multi-scale Feature and Multi-class of Fuzzy Mean Clustering[J].,2019,29(07):65-70.[doi:10. 3969 / j. issn. 1673-629X. 2019. 07. 013]
点击复制

多尺度多类中性模糊聚类图像分割算法()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
29
期数:
2019年07期
页码:
65-70
栏目:
智能、算法、系统工程
出版日期:
2019-07-10

文章信息/Info

Title:
Image Segmentation Algorithm Based on Multi-scale Feature and Multi-class of Fuzzy Mean Clustering
文章编号:
1673-629X(2019)07-0065-06
作者:
温金玉; 宣士斌; 黄亚武; 肖石林
广西民族大学 信息科学与工程学院,广西 南宁 530006
Author(s):
WEN Jin-yu; XUAN Shi-bin; HUANG Ya-wu; XIAO Shi-lin
School of Information Science and Engineering,Guangxi University for Nationlities, Nanning 530006,China
关键词:
中性聚类; 小波分析; 图像分割; 模糊聚类; 鲁棒性
Keywords:
neutrosophic clustering; wavelet analysis; image segmentation; fuzzy clustering; robustness
分类号:
TP37
DOI:
10. 3969 / j. issn. 1673-629X. 2019. 07. 013
摘要:
图像分割是将数字图像分成多个非重叠子区域的过程。 针对传统纹理分割方法在处理模糊边缘时容易产生边缘丢失,中性模糊聚类在处理图像时未考虑到局部空间信息,中性均值聚类模糊子集的聚类中心定义未充分考虑到每个样本点所属每个类的隶属度情况等问题,提出一种多尺度多类中性模糊聚类图像分割算法。 为了降低错分率和提高模糊退化的效率,使用确定性子集的数据分布来确定模糊子集的聚类中心。 为克服中性模糊聚类对噪声的敏感性,引入了小波去噪对图像进行预处理。 实验结果表明,所提出的分割方法的分割精确度高、有很强的抗噪能力,比起现有的聚类分割方法更能提高收敛速度。
Abstract:
Image segmentation is a process of dividing a digital images into several non-overlapping sub-regions. In order to solve the problem that the traditional texture segmentation easily generates edge loss when dealing with fuzzy edges,and the definition of clustering center that does not consider local spatial information, and neutrosophic mean clustering fuzzy subset does not fully consider the subordinate degree of each class belonging to each sample point when dealing with the image,a multi-scale and multi-class neutrosophic fuzzy clustering algorithm for image segmentation is proposed. In order to reduce the misclassification rate and improve the efficiency of fuzzy degradation,we use the data distribution of deterministic subset to determine the clustering center of fuzzy subset. In order to overcome the sensitivity of neutrosophic fuzzy clustering to noise,we introduce wavelet de-noising to image preprocessing. Experiment shows that the proposed segmentation method has high accuracy and strong noise immunity,and it can improve the convergence speed better than the existing clustering segmentation methods.

相似文献/References:

[1]王西锋 高岭 张晓孪.自相似网络流量预测的分析和研究[J].计算机技术与发展,2007,(11):42.
　WANG Xi-feng,GAO Ling,ZHANG Xiao-luan.Analysis and Research on Self-Similar Network Traffic Forecast[J].,2007,(07):42.
[2]淮文军 王明芳 汪梅[].基于小波分析的电缆故障特征提取方法研究[J].计算机技术与发展,2007,(11):209.
　HUAI Wen-jun,WANG Ming-fang,WANG Mei.Cable Fault Feature Extraction Method Research Based on Wavelet Analysis[J].,2007,(07):209.
[3]韩志艳,伦淑娴,王健.基于遗传小波神经网络的语音情感识别[J].计算机技术与发展,2013,(01):75.
　HAN Zhi-yan,LUN Shu-xian,WANG Jian.Speech Emotion Recognition Based on Genetic Wavelet Neural Network[J].,2013,(07):75.
[4]李圣普,王小辉,时合生.基于阈值的Mallat变换法的设计与仿真[J].计算机技术与发展,2014,24(04):107.
　LI Sheng-pu,WANG Xiao-hui,SHI He-sheng.Design and Simulation of Mallat Transform Method Based on Threshold[J].,2014,24(07):107.
[5]谈苗苗,成孝刚,周凯,等. 基于ARIMA和灰色模型加权组合的短期交通流预测[J].计算机技术与发展,2016,26(11):77.
　TAN Miao-miao,CHENG Xiao-gang,ZHOU Kai,et al. Short-term Traffic Flow Forecasting Based on Combination of ARIMA and Gray Model[J].,2016,26(07):77.
[6]杨祎玥[],伏潜[],万定生[]. 基于深度循环神经网络的时间序列预测模型[J].计算机技术与发展,2017,27(03):35.
　YANG Yi-yue[],FU Qian[],WAN Ding-sheng[]. A Prediction Model for Time Series Based on Deep Recurrent Neural Network[J].,2017,27(07):35.

更新日期/Last Update: 2019-07-10