• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>2016 5th International Conference on Informatics, Electronics and Vision >Differential evolution based optimal clustering for retinal blood vessel extraction
【24h】

Differential evolution based optimal clustering for retinal blood vessel extraction

机译：基于差异进化的视网膜血管提取最佳聚类

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Design of a computer-aided automatic system is very important for identification of different ocular diseases. A vital concern within this framework is the accurate retinal blood vessel extraction. This paper extracts vessels using curvelet transform, morphological operation, matched filtering and Differential Evolution based optimal clustering. Curvelet transform is implemented to enhance vessel edges. To remove the optic disc, the edge enhanced image is opened by a disk shaped structuring element which is then subtracted from the inverted histogram equalized image. Matched filtering intensifies the blood vessels response. The classification of the maximum matched filter responses into vessel and non-vessel classes is considered as a multi-objective clustering problem that minimizes the intracluster distances and maximizes the inter-cluster distance which is solved by Differential Evolution. Superiority of the proposed method is demonstrated by comparing it with the existing methods.
机译：计算机辅助自动系统的设计对于识别不同的眼部疾病非常重要。在此框架内，至关重要的问题是视网膜血管的准确提取。本文使用Curvelet变换，形态学运算，匹配滤波和基于差分演化的最佳聚类来提取血管。实施Curvelet变换以增强血管边缘。为了移除视盘，通过盘形结构元件打开边缘增强图像，然后从倒置的直方图均衡图像中减去该图像。匹配的过滤可增强血管反应。将最大匹配滤波器响应分为血管类和非容器类的问题被认为是一个多目标聚类问题，该问题可最小化集群内距离并最大化集群间距离，这可以通过差分演化解决。通过与现有方法进行比较证明了该方法的优越性。

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