多图像融合的连续腹腔影像图像分割  

Continuous abdominal images segmentation based on fusion images

作　　者：方艳红[1] 王梁[1] 吴斌[1] 

机构地区：[1]西南科技大学信息工程学院特殊环境机器人技术四川省重点实验室,四川绵阳621010

出　　处：《电子技术应用》2014年第8期126-128,共3页Application of Electronic Technique

基　　金：四川省教育厅项目(10ZA017);西南科技大学博士创新基金项目(13zx7113)

摘　　要：为实现连续腹腔影像图像分割的实时性和准确性,提出多图像融合的水平集图像分割模型。该模型通过Chan-Vese模型在预分割图像基础上获取形状信息,同时利用Li模型进一步在原始图像上获取边缘信息,以提取腹腔影像图中感兴趣区域;对相邻且变化缓慢的连续腹腔影像图,可将前一幅的分割结果作为下一幅的预分割图像,从而提高连续影像图像的分割效率。初步实验结果表明,该模型能实现目标区域相对连通的腹腔影像图像的有效分割,并且在处理连续腹腔影像图时处理效率较传统的方法有较大提高。To achieve a real-time and accurate segmentation of continuous abdominal images, an improved level set model composed with prior shape information in Chan-Vese and gradient information in Li model is proposed, which is used to extract the interesting region from abdominal image. To the trifle difference adjacent continuous abdominal images, the first image＇s seg- mentation can be the second image＇s pre-segmentation so as to improve the segmentation efficiency of the continuous abdominal images. Experiment results show that this model is able to effectively segment abdominal image from relatively connected region, and obviously improves time efficiency in segmenting continuous abdominal images.

关 键 词：连续腹腔影像图分割 区域生长法 水平集方法 Chan-Vese算法 

分 类 号：TP394.1[自动化与计算机技术—计算机应用技术] TH691.9[自动化与计算机技术—计算机科学与技术]