• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on Artificial Intelligence and Pattern Recognition >A Novel Segmentation Method for Microscopic Medical Images and Feature Extraction Using Region Adjacency Graph and automatic clustering method
【24h】

A Novel Segmentation Method for Microscopic Medical Images and Feature Extraction Using Region Adjacency Graph and automatic clustering method

机译：一种新的微观医学图像分段方法和使用区域邻接图和自动聚类方法的特征提取方法

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

In this paper, we propose a new unsupervised automatic segmentation method for microscopic medical images using the Region Adjacency Graph (RAG). First, Principle Component Analysis (PCA) is applied to the RGB color bands to enhance the contrast over the original image. Then, the RAG construction is obtained from a topological presegmentation based on watershed algorithm applied on the first PCA image. The RAG process is controlled by a novel statistical clustering method which combines the Generalized Likelihood Ratio (GLR) and the Bayesian Information Criterion (BIC). This approach yields the incorporation of the spectral characterization of the objects and their spatial relations. Structural features as cells area, shape indicator and cells color are extracted using the graph and then stored in a database in order to elaborate meaningful queries. Results show that our method that does not involve any a priori knowledge is suitable for several types of cytological images.
机译：在本文中，我们提出了一种使用区域邻接图（rag）的微观医学图像的新的无监督自动分段方法。首先，原理分量分析（PCA）应用于RGB色带，以增强原始图像上的对比度。然后，从基于在第一PCA图像上应用的流域算法的拓扑算法获得rag结构。 RAG过程由新颖的统计聚类方法控制，该方法结合了广义似然比（GLR）和贝叶斯信息标准（BIC）。该方法产生掺入对象的光谱表征及其空间关系。使用图形提取形状指示器和细胞颜色的结构特征，然后将其存储在数据库中以便详细说明有意义的查询。结果表明，我们不涉及任何先验知识的方法适用于几种类型的细胞学图像。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号