Journal of Computer Applications ›› 2011, Vol. 31 ›› Issue (05): 1217-1220.DOI: 10.3724/SP.J.1087.2011.01217

• Graphics and image technology • Previous Articles     Next Articles

Infrared small target detection algorithm based on improved two-sliding-window

LIU Xing-miao1, WANG Shi-cheng1, ZHAO Jing1,HU Bo2   

  1. 1. Accuracy Guidance and Control Laboratory, The Second Artillery Engineering College, Xi′an Shaanxi 710025, China
    2. Engineering Design and Research Institute of the Second Artillery, Beijing 100011, China
  • Received:2010-10-14 Revised:2010-12-03 Online:2011-05-01 Published:2011-05-01

基于改进双滑窗的红外小目标检测算法

刘兴淼1,王仕成1,赵静1,胡波2   

  1. 1.第二炮兵工程学院 精确制导与仿真实验室,西安710025
    2.第二炮兵工程设计研究院,北京100011
  • 通讯作者: 刘兴淼
  • 作者简介:刘兴淼(1981-),男,山东东明人,博士研究生,主要研究方向:红外图像处理、红外目标检测与跟踪;王仕成(1962-),男,山东单县人,教授,博士生导师,主要研究方向:导航、制导与控制、检测技术与自动化装置、图像处理、目标检测与跟踪;赵静(1981-),女,陕西西安人,博士研究生,主要研究方向:图形图像处理、碰撞检测。
  • 基金资助:

    中国博士后科学基金资助项目(20080441274);航空科学基金资助项目(20080112005)。

Abstract: The temporal domain characteristic of the infrared image and different features of small targets, noise and background were analyzed in this paper. A new infrared small target detection algorithm combining temporal domain and spatial domain was put forward. Because of the slow change of background, the Signal-to-Noise Ratio (SNR) of the small target was first enhanced through subtracting the conjoint frames. Then the potential small targets were detected by applying the centre distinguishing method, and the two-sliding-window algorithm was adopted to remove the isolated noise. At last, the similarity distinguishing method was used to eliminate the edge disturbance and the final detection of the small target was realized. The experimental results indicate that the improved algorithm has better target detection and real-time performance.

Key words: small target detection, two-sliding-window algorithm, infrared image, temporal domain characteristic, adaptive threshold

摘要: 分析了红外小目标图像的时域特性以及小目标、噪声、背景的不同特点,提出了一种时空结合的红外小目标检测算法。首先根据背景图像变化较慢的特点,运用相邻帧相减以减少背景和噪声的干扰,提高了目标信噪比(SNR);接着,使用中心点判别方法检测出可能的小目标点;然后,利用双滑窗方法去除孤立的噪声;最后,运用区域相似度判别函数,剔除边缘纹理的干扰,检测出小目标。仿真实验表明,该算法不仅具有良好的实时性,同时还具有较高的检测概率和较低的平均虚警数。

关键词: 小目标检测, 双滑窗算法, 红外图像, 时域特性, 自适应阈值