基于隐式模板超声图像序列的分割  

Implicit Template Based Segmentation of Ultrasound Image Sequences

作　　者：祝世海[1] 管伟光[1] 

机构地区：[1]哈尔滨理工大学计算机与控制学院,黑龙江哈尔滨150080

出　　处：《哈尔滨理工大学学报》2001年第3期18-20,共3页Journal of Harbin University of Science and Technology

摘　　要：研究了超声图像序列中形变结构的分割，根据隐函数对弹性结构的建模能力，提出了用隐式模板逼近形变结构的方法．待分割的结构是由一个能量最小化过程经过逼近而确定出来的．通过对超声心动描记图像进行实验的结果证明，本方法可有效地抑制超声图像序列中噪声对分割的影响，从而有利于对形变结构的提取。This paper makes an intensive research on the segmentation of deformed structures in ultrasound image sequences and presents the method of approximating the deformed structures with implicit template according to the modelling capability of implicit functions on elastic struc-tures. The structures which are to be segmented are determined by approximation using an energy minimization process. The experimental result of using this method on echocardiographic image sequences indicates that this method can effectively reduce the influence of noise on segmentation, so it can help to extract the deformed structures.

关 键 词：超声图像序列 分割 隐式模板 能量最小化 

分 类 号：TN911.73[电子电信—通信与信息系统]