Computer Engineering and Applications ›› 2012, Vol. 48 ›› Issue (21): 205-208.

Previous Articles     Next Articles

BEMD in weak and small target detection under complicated background

DENG Jianghua, WANG Ran, CHEN Zhong, CHENG Jiacong   

  1. National Key Lab of Science and Technology on Multi-Spectral Information Processing, Institute for Pattern Recognition and Artificial Intelligence, Huazhong University of Science and Technology, Wuhan 430074, China
  • Online:2012-07-21 Published:2014-05-19

BEMD在复杂背景下弱小目标检测中的应用

邓江华,王  然,陈  忠,程家聪   

  1. 华中科技大学 图像识别与人工智能研究所 多谱信息处理技术国家级重点实验室,武汉 430074

Abstract: For the purpose of weak and small target detection under complicated background, this paper proposes a target detection algorithm based on Bidimensional Empirical Mode Decomposition(BEMD). Images to be detected are decomposed into several bidimensional Intrinsic Mode Functions(IMF) as well as a residual image. Global threshold method is used to segment all IMFs. IMFs processed are synthesized to be one resulting image with segmented targets to be detected. The experiments have shown that the algorithm efficiently and accurately detects weak and small target with simplified method.

Key words: Bidimensional Empirical Mode Decomposition(BEMD), Intrinsic Mode Functions(IMF), weak and small target, segmentation, detection

摘要: 针对复杂背景下的弱小目标检测与识别问题,提出了一种基于二维经验模态分解(Bidimensional Empirical Mode Decomposition,BEMD)的检测算法。待检测的原图像经过BEMD分解筛选出多个二维的内蕴模函数(Intrinsic Mode Functions,IMF)和趋势图像,使用全局门限法分割各个IMF,将处理后的IMFs综合成一个分割出待检测目标的结果图像。实验结果表明,该方法使用简洁的步骤,有效、准确地检测出弱小目标。

关键词: 二维经验模态分解, 内蕴模函数, 弱小目标, 分割, 检测