摘要
采用基于各向异性扩散的偏微分方程,其初始值为输入图像,转化为差分格式迭代求解滤波结果。在去除噪声的同时,保持重要的边缘和局部细节。在此基础上提出了8向的各向异性扩散和边缘增强的处理技术,取得了满意的结果,并将此切片图像经聚类分群运用到三维重构中,使重构的效果更好。
This paper studied the way to remove the speckle noise in medical images by using anisotropic diffusion method. Based on anisotropic diffusion, a partial differential equation, of which the initial data was the input images, was transformed into differential forms and solved with iterations. Anisotropic diffusion can remove the speckle noise efficiently and preserve edges and local detail at the same time. The processed slices image would be further grouped by clustering and be reconstructed efficiently.
出处
《计算机应用》
CSCD
北大核心
2007年第1期249-251,共3页
journal of Computer Applications
基金
江西省教育厅2004年科技基金资助项目(104012)
关键词
偏微分方程
各向异性扩散
图像分割
模糊聚类
三维重构技术
partial differential equation
anisotropic diffusion
image segmentation
fuzzy clustering
3D-reconstruction technology
