J4 ›› 2009, Vol. 36 ›› Issue (6): 1075-1078.

• 研究论文 • 上一篇    下一篇

基于背景预测的红外弱小目标检测新算法

李凡;刘上乾;洪鸣;秦翰林   

  1. (西安电子科技大学 技术物理学院,陕西 西安  710071)
  • 收稿日期:2009-03-10 出版日期:2009-12-20 发布日期:2010-01-20
  • 通讯作者: 李凡
  • 作者简介:李凡(1985-),女,西安电子科技大学博士研究生,E-mail: fionalucifer@gmail.com.
  • 基金资助:

    国家部委预研资助(41101050102)

Dim infrared targets detection based on background prediction

LI Fan;LIU Shang-qian;HONG Ming;QIN Han-lin   

  1. (School of Technical Physics, Xidian Univ., Xi'an  710071, China)
  • Received:2009-03-10 Online:2009-12-20 Published:2010-01-20
  • Contact: LI Fan

摘要:

针对复杂背景中弱小目标检测难的问题,提出一种改进的双边滤波背景预测算法.该算法在双边滤波中引入梯度算子,自适应地对背景进行预测,将原始图像与预测图像相减可以抑制背景细节、增强目标信息,同时利用梯度的统计特性减少算法的计算量,提高了弱小目标检测性能.仿真和实验表明,与双边滤波的检测算法相比,该算法能够更加有效地从复杂背景中检测弱小目标.

关键词: 图像处理, 背景预测, 弱小目标检测, 红外图像

Abstract:

In order to solve the difficult problem of dim targets detection in complicated background, a new bilateral filtering method for background prediction is presented. In this method, the true image background is estimated adaptively via bilateral filtering, and the estimated image is subtracted from the source image to suppress background details and enhance target information. To improve the performance of dim targets detection, the gradient statistics is introduced to decrease the compatational amount. Simulations and experiments show that the new algorithm is more efficient in detecting the dim target from complicated background compared with the method based on bilateral filtering.

Key words: image procession, background prediction, dim targets detection, infrared imaging

中图分类号: 

  • TP391
Baidu
map