结合区域信息的分段活动轮廓模型  

Segmental Active Contour Model Combining Regional Information

作　　者：冉鑫[1,2] 戚飞虎[1] 

机构地区：[1]上海交通大学计算机科学与工程系 [2]上海海事大学商船学院,上海200135

出　　处：《生物医学工程学杂志》2006年第6期1166-1171,共6页Journal of Biomedical Engineering

基　　金：国家自然科学基金资助项目(60271033)

摘　　要：提出了一种结合区域信息的分段活动轮廓模型,利用边缘信息迅速找到对象的大体轮廓,然后结合区域统计信息使模型精确收敛到对象边缘。分段的层次化变形有效的利用了图像的全局和局部信息,使用仿射变换使模型的局部以同一种变换方式变形,提高模型对噪声和伪边缘的鲁棒性,同时保持模型轮廓形状的一致性。在精确匹配阶段利用区域统计信息重新定义模型的外部能量,采用自适应的搜索区域确定方法,提高了算法的效率和进入凹边缘的能力。试验表明本模型运算速度快,抗噪声和避免陷入局部极小值的能力较强,有较好的分割效果。A segmental active contour model integrating region information is proposed. Different deformation schemes are used at two stages for segmenting the object correctly in image plane. At the first stage the contour of the model is divided hierarchically into several segments, each of which deforms respectively using affine transformation. After the contour deforms to the approximate boundary of object, a fine matching method using statistical information of Local region to redefine the external energy of the model is used to make the contour fit the object＇s boundary exactly. The algorithm is effective, as the reformative approaches of computing the internal energy and external energy are proposed to reduce the algorithm complexity. The experimental results indicate that the proposed model is robust to local minima and able to search for concave objects.

关 键 词：活动轮廓模型 图像分割 区域增长 医学图像处理 

分 类 号：R318[医药卫生—生物医学工程]