文章摘要
李传龙,李颖,刘爱莲.灰度不均匀图像分割[J].,2014,54(1):106-114
灰度不均匀图像分割
Segmentation of image with intensity inhomogeneity
  
DOI:10.7511/dllgxb201401017
中文关键词: 灰度不均匀  活动轮廓  LBF模型  图像分割  窄带
英文关键词: intensity inhomogeneity  active contour  local binary fitting (LBF) model  image segmentation  narrow band
基金项目:国家自然科学基金资助项目(41171329,41071260);“十一五”国家科技支撑计划资助项目(2006BAC11B01);大连海事大学基本科研业务费资助项目.
作者单位
李传龙,李颖,刘爱莲  
摘要点击次数: 1806
全文下载次数: 6807
中文摘要:
      采用虚拟的符号距离函数代替真实的符号距离函数,依靠待检测目标局部灰度高斯加权均值来驱动活动轮廓的演化,提出了一种能够分割灰度不均匀图像的新颖活动轮廓模型.利用虚拟符号距离函数的梯度形成一个窄带,活动轮廓在窄带内做演化运算,其演化具有计算简单、分割效率高、抗噪性强等优点.符号距离函数重新初始化也只需要在窄带内使用高斯函数规则化后,对其取符号运算即可.符号距离函数重新初始化具有计算简单、效率高的特点.最后给出了活动轮廓在窄带内收敛的一个简单条件,能方便地判断待检测目标是否被检测出来.
英文摘要:
      By replacing the real symbol distance function with the virtual symbol distance function, and depending on the Gaussian weighted averages of the local region of the object to drive the active contour evolution, a novel active contour model is presented for the segmentation of the image with intensity inhomogeneity. The gradient of the virtual symbol distance function forms a narrow band, where the active contour evolutes by simple calculation. Thus, the evolution has the advantages of simple calculation, high-efficiency segmentation, strong anti-noise nature, etc.. The virtual symbol distance function is re-initialized with sign function after the virtual symbol distance function is regularized by the Gaussian function within the narrow band. The re-initialization in the presented model is simple and efficient. In addition, a simple condition is given for the active contour convergence within the narrow band to judge whether the object is detected or not.
查看全文   查看/发表评论  下载PDF阅读器
关闭