采用信息融合的红外目标检测算法  

Infrared target detection algorithm based on information fusion

作　　者：郑武兴 王春平 付强 徐艳 李宁 ZHENG Wu-xing;WANG Chun-ping;FU Qiang;XU Yan;LI Ning(Department of Electrics and Optics Engineering,Shijiazhuang Campus of Army Engineering University,Shijiazhuang 050003,China;Department of Teaching and Research,Army Artillery Training Base,Zhangjiakou 075100,China)

机构地区：[1]陆军工程大学石家庄校区电子与光学工程系,河北石家庄050003 [2]陆军炮兵训练基地教研部,河北张家口075100

出　　处：《激光与红外》2018年第2期261-266,共6页Laser & Infrared

基　　金：国家自然科学基金项目(No.61141009)资助

摘　　要：针对空中红外目标检测过程中存在的检测精度不高、易受干扰等问题,本文提出了一种信息融合的目标检测算法。首先,针对单一滤波模式的不足,采用降采样方式将多种模式的滤波方法进行融合,有效提高了滤波后的图像信噪比,减少了预处理时间。然后,对质心检测选定的检测区域进行显著性检测和边缘检测,并将结果进行融合,增强了目标区域的信号强度,改善了算法的检测精度。文中从理论上介绍和分析了该算法对红外目标检测的有效性,并通过Matlab仿真实验与其他单一检测算法进行了对比。实验结果表明,信息融合的检测算法能够在多种复杂情况下有效检测出空中红外目标,证明融合检测算法的抗干扰性和适应性上更强、检测精度更高、算法的鲁棒性更好。As aerial infrared target detection has low detection accuracy and is susceptible to interference,a target de-tection algorithm based on information fusion is proposed.Firstly,the filtering method of multiple modes was fused by downsampling to effectively improve the SNR of filtered image and reduce the preprocessing time.Then,the saliency detection and edge detection were carried out for selected detection area of centroid detection,and the results were fused to enhance signal intensity of target area and improve detection precision.The effectiveness of the proposed al-gorithm was theoretically introduced and analyzed,and the Matlab simulation results were compared with that of other single detection algorithms.The experimental results show that the detection algorithm of information fusion can effec-tively detect aerial infrared target in a variety of complex situations.It is proved that the fusion detection algorithm has stronger anti-jamming and adaptability,higher detection precision and better robustness.

关 键 词：信息融合 红外目标 融合滤波 目标检测 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]