摘要
为实现CT图片自动准确分割,提出一种基于Gabor变换和加权欧式距离模糊C均值的图像分割算法.该算法使用Gabor变换提取图像特征组成特征向量作为图像分割依据,并基于加权欧式距离FCM算法对图像进行分割,权值通过有监督的网格搜索方法进行优化.对脑部CT图片进行分割实验,结果表明,相对于FCM和PCM算法,本文提出的算法分割结果噪声更少,边缘更加清晰,证明该算法的有效性.
In order to achieve automatic and accurate CT image segmentation,this paper proposes an image segmentation algorithm based on Gabor transform and weighted fuzzy C-means(FCM).Gabor transform is used to extract image features,and the appropriate feature vectors are selected as the basis of image segmen⁃tation.Then a weighted Euclidean distance FCM algorithm is used to segment the image,in which the weight is optimized by supervised grid search method.Compared with FCM and PCM algorithm,the algorithm pro⁃posed in this paper has less noise and clearer edge,which proves the effectiveness of the algorithm.
作者
赵娟
袁慧宇
ZHAO Juan;YUAN Huiyu(School of Computer Science and Technology,Huaibei Normal University,235000,Huaibei,Anhui,China;Information College,Huaibei Normal University,235000,Huaibei,Anhui,China)
出处
《淮北师范大学学报(自然科学版)》
CAS
2020年第4期58-63,共6页
Journal of Huaibei Normal University:Natural Sciences
基金
安徽省高校自然科学项目(KJ2017B017)
2018年高校优秀青年人才支持计划(gxyq2018161)。