计算机应用 ›› 2011, Vol. 31 ›› Issue (08): 2249-2252.DOI: 10.3724/SP.J.1087.2011.02249

• 图形图像技术 • 上一篇    下一篇

基于图像块分类器和条件随机场的显微图像分割

阳维1,2,张树恒2,王莲芸3,张素2   

  1. 1. 南方医科大学 生物医学工程学院,广州510515
    2. 上海交通大学 生物医学工程学院,上海200240
    3. 上海交通大学 生命科学技术学院,上海200240
  • 收稿日期:2011-02-24 修回日期:2011-04-21 发布日期:2011-08-01 出版日期:2011-08-01
  • 通讯作者: 张素
  • 作者简介:阳维(1979-),男,湖北天门人,讲师,博士,主要研究方向:图像处理、模式识别;张树恒(1988-),男,河北邯郸人,硕士研究生,主要研究方向:生物医学图像处理;王莲芸(1957-),女,山西阳泉人,教授,主要研究方向:环境因素对免疫功能的影响;张素(1968-),女,江苏无锡人,副教授,主要研究方向:图像处理。

Segmentation of microscopic images based on image patch classifier and conditional random field

Wei YANG1,2,Shu-heng ZHANG1,Lian-yun WANG3,Su ZHANG1   

  1. 1. School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China
    2. School of Biomedical Engineering, Southern Medical University, Guangzhou Guangdong 510515, China
    3. School of Life Science and Biotechnology, Shanghai Jiao Tong University, Shanghai 200240, China
  • Received:2011-02-24 Revised:2011-04-21 Online:2011-08-01 Published:2011-08-01
  • Contact: Su ZHANG

摘要: 针对花粉显微图像处理提出了一种自动分割方法,将有助于花粉识别系统的开发。使用归一化颜色分量训练图像块分类器,并且结合条件随机场和图割进行建模和优化,利用最大化后验概率(MAP)的方法实现花粉显微图像中花粉区域的分割。对于实验中的133幅图像,自动分割同人工分割的结果相比较,统计得到距离误差均值为7.3像素,准确率的平均值为87%。实验结果表明,使用图像块分类器和条件随机场模型可以用于花粉图像的分割。

关键词: 花粉显微图像, 图像分割, 图像块分类器, 条件随机场, 图割

Abstract: An automatic segmentation for pollen microscopic images was proposed in this paper, which was useful to develop a recognition system of airborne pollen. First, the image patch classifier was trained with normalized color component. Then, conditional random field was employed to model pollen images and Maximum A Posterior (MAP) was used to segment the pollen areas in microscopic images, with graph cut algorithm for optimization. In the experiments, the respective average values of mean distance error was 7.3 pixels and the true positive rate was 87% on 133 images. The experimental results show that image patch classifier and conditional random field model can be used to accomplish segmentation of the pollen microscopic images.

Key words: pollen microscopic image, image segmentation, image patch classifier, Conditional Random Field (CRF), graph cut

中图分类号: