基于电势能改进区域生长脑肿瘤图像分割.PDF

第21卷第8期 Vol.21 No.8 2018年8月 软件工程 SOFTWARE ENGINEERING Aug. 2018 文章编号：2096-1472(2018)-08-01-03 DOI:10.19644/ki.issn2096-1472.2018.08.001 基于电势能改进的区域生长脑肿瘤图像分割 蒋秋霖，王 昕 (长春工业大学计算机科学与工程学院，吉林 长春 130012) 摘 要：脑肿瘤的图像分割被广泛应用于临床诊断中。为提高脑肿瘤分割的精确度，本文提出了一种结合电势能 的区域生长算法分割MR脑肿瘤图像。将图像的每个像素点看作是电荷，像素值作为电荷量，肿瘤区域像素值大因而其 “电荷量”大于其他非肿瘤区域“电荷量”，因此建立生长准则。实验对比其他两种典型算法，本文算法分割精度最优。 关键词：脑肿瘤分割；电势能；区域生长 中图分类号：TP391.41 文献标识码：A Image Segmentation of Region Growing Brain Tumor Based on Potential Energy Segmentation JIANG Qiulin,WANG Xin ( School of Computer Science Engineering ,Changchun University of Technology ,Changchun 130012,China) Abstract:Image segmentation of brain tumors is widely used in clinical diagnosis.In order to improve the accuracy of brain tumor segmentation,this paper proposes a region growing algorithm based on potential energy segmentation to segment MR brain tumor images.Each pixel of the image is regarded as a charge,and the pixel value is used as the charge amount. Since the tumor area has a larger pixel and thus its charge amount is larger than that of other non-tumor areas,a growth criterion is established.Compared two other typical algorithms,the proposed algorithm has the best segmentation accuracy. Keywords:brain tumor segmentation;electric potential energy;region growing 1 引言(Introduction) 确。Weglinski等[6]先通过中值滤波减少噪声，再利用区域生 目前，脑肿瘤图像分割已经成为一项众多学者研究的课 长法分割脑肿瘤图像。Deng等[7]在传统区域生长法的基础上， 题，通过