• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International conference on sensing and imaging >Phase Congruency and Its Application to Tubular Structure Extraction
【24h】

Phase Congruency and Its Application to Tubular Structure Extraction

机译：相同时及其在管状结构提取的应用

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

摘要

A number of tubular structure extraction algorithms have been developed in the literature. Thresholding [1] is the fastest one among all of them. For images from the industry, however, the varied contrast of the tubular structures makes it difficult to determine an appropriate threshold. To tackle this issue, noise suppression and tubular structure enhancement are usually applied before thresholding. Popular enhancement methods are based on filtering with filters constructed from an analysis of multi-scale space or local Hessian matrix, such as the Frangi operator [2] and Krissian operator [3]. The response of the Hessian matrix-based operators is considerably sensitive to the local image contrast. In order to overcome the shortcomings of Frangi and Krissian operators, Bauer [4] proposed an approach based on gradient vector flow (GVF) fields. By providing a new external force for active contours and snakes, the GVF method has a strong ability to suppress noise. However, it will lose some of fragile tubular structures. The recently proposed method [5, 6] defines a nonlinear and nonlocal path operators, which can be used to filter out tubular structures. However, its low efficiency prevents its further application to real CT images which are often big volume datasets.
机译：在文献中开发了许多管状结构提取算法。阈值化[1]是所有中最快的一个。然而，对于行业的图像，管状结构的变化对比度使得难以确定适当的阈值。为了解决这个问题，通常在阈值化之前施加噪声抑制和管状结构增强。流行的增强方法基于滤波，通过分析多尺度空间或本地Hessian矩阵的分析，例如弗朗尼运算符[2]和克里斯安运算符[3]。对基于矩阵的运算符的响应对本地图像对比度相当敏感。为了克服弗朗尼和克里斯安运营商的缺点，鲍尔[4]提出了一种基于梯度矢量流（GVF）字段的方法。通过为活动轮廓和蛇提供新的外力，GVF方法具有强大的抑制噪声的能力。然而，它将失去一些脆弱的管状结构。最近提出的方法[5,6]限定了非线性和非本地路径运算符，可用于过滤输出管状结构。然而，其低效率可防止其进一步应用于通常是大卷数据集的真实CT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