基于数学形态学的差分图像目标检测算法研究  

Differencing Image Target Detection Algorithm Based on Mathematical Morphology

作　　者：秦雨萍[1,2] 张双 张萍[1] 尹福成 QIN Yuping;ZHANG Shuang;ZHANG Ping;YIN Fucheng(Engineering&Technical College,Chengdu University of Technology,Leshan 614000;Neijiang Normal University,Neijiang 641000)

机构地区：[1]成都理工大学工程技术学院,乐山614000 [2]内江师范学院,内江641000

出　　处：《舰船电子工程》2018年第4期16-18,22,共4页Ship Electronic Engineering

基　　金：国家高技术研究发展计划(863计划)基金(编号:A08K021);国家自然科学基金面上项目(编号:61201397;11375055);四川省教育厅重大培育项目(编号:17CZ005);四川省教育厅基金项目(编号:17ZA047);成都理工大学工程技术学院青年基金(编号:C122016030)资助

摘　　要：针对舰船运动目标跟踪瞄准中,目标受背景影响比较严重这一问题,文章提出将通过运动补偿实现图像的配准后的图像,通过差分运算,以及利用多帧差分图像序列,获得平均差分图像,从而提高信噪比,系统设计为了减小噪声点对目标检测的干扰,利用自适应阈值对差分图像二值化,对二值化后的图像用形态学处理消除噪声斑点,最后通过连通区域像素个数统计判决,进一步去除伪目标,最终检测出目标,通过实验验证该方法是有效可行的。In tracking and pointing of moving targets on the vessel,the background has a great effect on targets.Therefore,an implement image registration is proposed through motion compensation,and the mean differencing images can be obtained through differential operation and by means of multi-frame differencing image sequence,so as to enhance signal to noise ratio.In the system design,to reduce interference of noise points upon target detection,differencing images are binarized by using the adaptive threshold and noise spots of the binarized images are processed using the morphology,finally,false targets are further eliminated by statistical decision of pixel number in the connected region.Ultimately the target is detected.Experimental verification demonstrates that the method is effective and feasible.

关 键 词：目标跟踪 差分运算 数学形态学 目标检测 

分 类 号：TN911.73[电子电信—通信与信息系统]