摘要
把肝脏及其内部各种管道从CT序列图像中分割出来,为后续肝脏及其管道的三维重建及最终的虚拟手术模拟提供正确的数据。在结合CT序列图像之间的相似性基础之上,提出一种动态自适应的区域生长算法。首先对原始CT数据进行中值滤波,去除部分噪声,然后采用相应的序列化区域生长分割模型,对CT序列图像中肝脏及其内部管道分别进行提取。实验结果表明,应用该方法能得到准确的肝脏及管道分割结果。
Extracting the liver and its intern vessels from CT images can provide the correct data for the following 3D reconstruction of liver and the final virtual surgery.Based on the similarity between CT sequence images,an algorithm of auto-adapted region growing was put forward.First,originality CT data have been reduced noise by mean-filter,then based on the serializing division model of region growing,the liver and its vessels were segmented from CT images.The results of experiment show that the method is effective and very useful in automatic segmentation of CT sequence image.
出处
《计算机工程与应用》
CSCD
北大核心
2009年第20期205-207,245,共4页
Computer Engineering and Applications
基金
国家高技术研究发展计划(863)No.2006AA02Z346
广东省自然科学基金团队项目(No.6200171)~~
关键词
序列图像
相似性
自适应
区域生长
sequence image
similarity
auto-adapted
region-growing