作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2007, Vol. 33 ›› Issue (14): 168-170. doi: 10.3969/j.issn.1000-3428.2007.14.059

• 人工智能及识别技术 • 上一篇    下一篇

空间遮挡点目标检测的新方法

张海英1,张田文2,温 玄3   

  1. (1. 厦门大学软件学院,厦门 361005;2. 哈尔滨工业大学计算机科学与技术学院,哈尔滨 150001;3. 厦门大学嘉庚学院,厦门 363105)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-07-20 发布日期:2007-07-20

New Method for Detection of Point Targets with Occlusion in Space

ZHANG Haiying1, ZHANG Tianwen 2, WEN Xuan3   

  1. (1. School of Software, Xiamen University, Xiamen 361005; 2. School of Computer Science and Technology, Harbin Institute of Technology, Harbin 150001; 3. School of Jiageng, Xiamen University, Xiamen 363105)
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-07-20 Published:2007-07-20

摘要: 为了处理云层遮挡下的点目标检测问题,遵循边检测边跟踪的理论框架,该文设计了一种基于灰度与对比度信息融合的检测方法。通过检测累积的目标轨迹完成复杂情况下的多目标检测,后续的实验表明了该方法的有效性。

关键词: 点目标, 轨迹互联, 局部对比度, 边检测边跟踪

Abstract: In order to detect the point targets in the case of occlusion with cloud, following the track before detection framework, this paper presents an information association method based on grey and contrast. Multiple targets detection is achieved based on the accumulated trajectories. The following experiments prove its validity.

Key words: point targets, trajectories association, local contrast, track before detection

中图分类号: