• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Target-in-the-Loop: Atmospheric Tracking, Imaging, and Compensation >TrackEye tracking algorithm characterization
【24h】

TrackEye tracking algorithm characterization

机译：TrackEye跟踪算法特征

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

TrackEye is a film digitization and target tracking system that offers the potential for quantitatively measuring the dynamic state variables (e.g., absolute and relative position, orientation, linear and angular velocity/acceleration, spin rate, trajectory, angle of attack, etc.) for moving objects using captured single or dual view image sequences. At the heart of the system is a set of tracking algorithms that automatically find and quantify the location of user selected image details such as natural test article features or passive fiducials that have been applied to cooperative test articles. This image position data is converted into real world coordinates and rates with user specified information such as the image scale and frame rate. Though tracking methods such as correlation algorithms are typically robust by nature, the accuracy and suitability of each TrackEye tracking algorithm is in general unknown even under good imaging conditions. The challenges of optimal algorithm selection and algorithm performance/measurement uncertainty are even more significant for long range tracking of high-speed targets where temporally varying atmospheric effects degrade the imagery. This paper will present the preliminary results from a controlled test sequence used to characterize the performance of the TrackEye tracking algorithm suite.
机译：TrackEye是一种胶片数字化和目标跟踪系统，它可以定量测量动态状态变量（例如，绝对和相对位置，方向，线速度和角速度/加速度，旋转速度，轨迹，攻角等）。使用捕获的单视图或双视图图像序列移动物体。系统的核心是一组跟踪算法，这些算法可自动查找和量化用户选择的图像细节的位置，例如自然色的测试品特征或已应用于协作测试品的被动基准。使用用户指定的信息（例如图像比例和帧速率）将此图像位置数据转换为现实世界的坐标和速率。尽管诸如相关性算法之类的跟踪方法通常本质上都是健壮的，但是即使在良好的成像条件下，每个TrackEye跟踪算法的准确性和适用性通常也是未知的。最佳算法选择和算法性能/测量不确定性的挑战对高速目标的远距离跟踪更为重要，因为在这种情况下，随着时间变化的大气影响会降低图像质量。本文将提供来自受控测试序列的初步结果，这些序列用于表征TrackEye跟踪算法套件的性能。

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