背景约束的红外复杂背景下坦克目标分割方法  

Tank Segmentation Under Infrared Complex Background with Background Restriction

作　　者：高敏[1] 李怀胜 周玉龙[1,3] 方丹[1] 张宝全[4] 

机构地区：[1]军械工程学院导弹工程系 [2]66393部队装备部 [3]66393部队博士后科研工作站 [4]国网河北省电力公司检修分公司

出　　处：《自动化学报》2016年第3期416-430,共15页Acta Automatica Sinica

基　　金：中国博士后科学基金(2014M562657)资助~~

摘　　要：为实现自寻的反坦克导弹红外导引头对复杂背景下坦克目标的快速有效分割,以红外导引头拍摄的坦克目标红外图像为研究对象,采用计算简单的最大类间方差法,对其分割效果进行了研究.根据实验结果,揭示了最大类间方差法进行图像分割的有效性机理.在此基础上,提出了对背景区域像素和灰度级别进行约束的思想,以降低背景区域类内方差,提高算法的分割精度,并给出了具体的方法.首先利用坦克目标的先验信息,根据光学成像原理,推导了红外坦克目标图像的大小估计公式,用来实现对背景像素的约束;然后采用黄金分割法对背景灰度级别进行约束;最后利用最大类间方差法实现了复杂背景下红外坦克目标的分割.实验表明,本文方法的分割效果堪比手工分割效果,且计算量较少,算法耗时最大不超过1.44 ms,完全满足对坦克目标图像分割的有效性和实时性需求.To achieve fast and efficient tank segmentation under infrared complex background for infrared imaging guidance system of homing antitank missile, the maximum between-class variance method is studied using tank infrared images from the infrared imaging and guidance system, and the working mechanism of the maximum between-class variance method is revealed. Then the idea to restrict the background pixels and gray levels is proposed to lower the background within-class variance so as to improve the segmentation accuracy of the method. The detailed approach is as follows. Firstly, with the prior information of the tank, according to the optical imaging principle, the size equation of the tank image is deduced, by which background pixels are restricted; secondly, the golden section is used to restrict the background gray levels; lastly, with the maximum between-class variance method, segmentation of the tank target is achieved. Experiments show that the segmentation result of the proposed method is as good as manual segmentation s and the processing time is decreased to no more than 1.44 ms. So it can completely meet the needs of effectiveness and real-time processing for the tank segmentation under the infrared complex background of the homing antitank missile.

关 键 词：自寻的反坦克导弹红外成像制导系统 坦克目标分割 最大类间方差法 阈值分析 背景约束 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]