首页> 外文会议>European Association of Remote Sensing Laboratories Symposium >Application of mathematical morphology operations for the improvement of identification of linear objects preliminarily extracted from classification of VHR satellite images
【24h】

Application of mathematical morphology operations for the improvement of identification of linear objects preliminarily extracted from classification of VHR satellite images

机译:数学形态学作用在VHR卫星图像分类中提取识别线性对象的识别

获取原文

摘要

In this paper the method of extracting and reconstruction of image of linear object on VHR satellite image on the example of road is presented. The method is based on simple operators of mathematical morphology while most of algorithms with the same purpose are linear. The first step of approach presented is an extracting of the mask of the object using traditional classification algorithms. Another step which is a point of the algorithm is a sequence of morphological operations. The study shows how important for the result is the shape of structuring elements used in the operations.
机译:本文提出了道路示例的VHR卫星图像上线性对象图像的提取和重建方法。该方法基于数​​学形态的简单操作者,而具有相同目的的大多数算法是线性的。呈现的第一步是使用传统分类算法提取物体的掩模。另一个步骤是算法的点是一系列形态操作。该研究表明,结果是在操作中使用的结构元件的形状。

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