• 主题
高级搜索  >
历史搜索 清空历史
首页> 美国政府科技报告 >Anvaendning av Datafusion vid Klassificering av Markmal, en Inledande Studie(Usage of Datafusion for Classification of Ground Targets, A Preliminary Study).
【24h】

Anvaendning av Datafusion vid Klassificering av Markmal, en Inledande Studie(Usage of Datafusion for Classification of Ground Targets, A Preliminary Study).

机译：anvaendning av Datafusion vid Klassificering av markmal，en Inledande studie（Datafusion用于地面目标分类，初步研究）。

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

摘要

This study investigates how datafusion may be used to classify objects which have been observed and tracked for some time. During tracking some features of the target have been measured and revised within the resolution capabilities of the sensors. The classification is then done by using these object features. Under some assumptions, a simplified simulation environment for generating silhouette-based features has been constructed. The simulation environment uses geometric models for some target types for generation of the silhouettes. Depending on these silhouettes, the distributions for the selected features have been generated and the classification is then made by using these distributions. The method used for classification is Bayes decision rule. Some of the conclusions are that the use of sensors with complementary properties gives improved and more robust classification. Classification with the use of datafusion gives a better separation between the different classes which leads to less severe classification errors.

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