J4 ›› 2012, Vol. 39 ›› Issue (3): 95-99+130.doi: 10.3969/j.issn.1001-2400.2012.03.015

• 研究论文 • 上一篇    下一篇

边缘结构背景杂波尺度

李倩;张建奇;杨翠   

  1. (西安电子科技大学 技术物理学院,陕西 西安  710071)
  • 收稿日期:2011-09-06 出版日期:2012-06-20 发布日期:2012-07-03
  • 通讯作者: 李倩
  • 作者简介:李倩(1985-),女,西安电子科技大学博士研究生,E-mail: qianli@mail.xidian.edu.cn.
  • 基金资助:

    国家自然科学基金资助项目(60777042);中央高校基本科研业务费资助项目(K50510050004,K50510050001)

Structure of the edge background clutter metric

LI Qian;ZHANG Jianqi;YANG Cui   

  1. (School of Technical Physics, Xidian Univ., Xi'an  710071, China)
  • Received:2011-09-06 Online:2012-06-20 Published:2012-07-03
  • Contact: LI Qian

摘要:

根据观察主体主要依赖结构特征识别目标的事实,以背景和目标边缘结构分布的差别来表征二者之间结构特征的相似性,建立了边缘结构背景杂波尺度.与以往的杂波尺度相比,该尺度通过统计目标和背景结构的空间分布特性,强调物体自身结构的重要性,削弱环境亮度对目标识别的影响,适用于各种自然场景.实验数据表明,提出的杂波尺度与现场目标获取性能有很好的一致性.

关键词: 目标获取性能, 目标识别, 背景杂波, 结构特征, 边缘方向分布

Abstract:

Based on the fact that the recognition of a target depends mainly on the structure feature for human vision, the structure of the edge background clutter metric is proposed, by describing the similarity between background and target by their difference in the distribution of the edge structure. Compared to the others, it is available for various scenes by statistically computing the properties of the target and background's structure to emphasize the importance of the structure and weaken the influence of scene illumination on target recognition. Experiment shows that its prediction correlates well with the detection probability of human visual systems.

Key words: target acquisition performance, target recognition, background clutter, structure feature, distribution of edge orientation

中图分类号: 

  • TN216
Baidu
map