...
【24h】

Wide-Area Motion Imagery

机译:广域动态影像

获取原文
获取原文并翻译 | 示例
           

摘要

Wide-area motion imagery (WAMI) sensors are placed on helicopters, balloons, small aircraft, or unmanned aerial vehicles and are used to image small city-sized areas at approximately 0.5 m/pixel and about one or two frames/s. The geospatial-temporal data sets produced by these systems allow for the observation of many dynamic phenomena that were previously inaccessible in street-level video data, but the efficient exploitation of this data poses significant technical challenges for image and video analysis and for data mining. Content of interest is defined in very abstract terms related to how humans interpret video imagery, but the data is defined in very physical terms related to the imaging device. This difference in representations is often called the semantic gap. In this review article, we describe advances that have been made and the advances that will be needed to produce the hierarchy of computational models required to narrow the semantic gap in WAMI.
机译:广域运动图像(WAMI)传感器放置在直升机,气球,小型飞机或无人飞行器上,用于以约0.5 m /像素和约一或两帧/秒的速度拍摄小城市区域。这些系统产生的地理时空数据集允许观察以前在街道级视频数据中无法访问的许多动态现象,但是如何有效利用这些数据对图像和视频分析以及数据挖掘提出了重大的技术挑战。感兴趣的内容以与人类如何解释视频图像有关的非常抽象的术语定义,但是数据以与成像设备有关的非常物理的术语定义。这种表示上的差异通常称为语义鸿沟。在这篇评论文章中,我们描述了已经取得的进步以及产生缩小WAMI中语义差距所需的计算模型层次结构所需的进步。

著录项

  • 来源
    《Signal Processing Magazine, IEEE》 |2010年第5期|P.56-65|共10页
  • 作者

  • 作者单位

    Research scientist at Los Alamos National Laboratory;

  • 收录信息
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

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