基于梯度分析的弹丸检测算法  

Research on Projectile Detection Algorithm Based on Gradient Analysis

作　　者：邹斌 张斌[1] 褚文博 孟凡军[2] 张大舜[2] 赵冬娥[1] ZOU Bin;ZHANG Bin;CHU Wenbo;MENG Fanjun;ZHANG Dashun;ZHAO Donge(School of information and communication engineering, North University of China, Taiyuan 030051, China;The Fifth Research Institute of China Ordnance Industry Group, Changchun 130000, China)

机构地区：[1]中北大学电子测试技术国防重点实验室,太原030051 [2]中国兵器工业集团第五五研究所,长春130000

出　　处：《兵器装备工程学报》2021年第3期258-262,共5页Journal of Ordnance Equipment Engineering

摘　　要：针对CCD交汇测坐标系统测试武器时可能出现外界干扰导致弹丸漏检的问题,提出了一种基于背景差分法和梯度分析的目标检测算法。首先利用线阵相机单幅图像由多行叠加的特性,将图像在纵向上翻转再与原图相减,消除大部分静态背景得到预处理图像;然后对其进行梯度矢量检测,此时弹丸相比于背景噪声在矢量图上的幅值更大;再将矢量图二值化后对其进行Hough直线检测,选取合适的角度和幅度等参数,得到弹丸在图像坐标系上的像素坐标;最后经过坐标变换得到弹丸在靶面上的过靶坐标。与其他目标检测算法进行对比,该算法在保证检测速度的前提下,能够有效提高弹丸检测准确率达95%以上,适用性强,可用于多类武器测试。In order to solve the problem of missing projectile detection caused by external interference in CCD intersection coordinate measuring system,a target detection algorithm based on background difference method and gradient analysis was proposed.In this case,the background image was first detected by using the gradient value of the original line image,and then the image was detected by superimposing a large number of line images on the background vector image,and then the background image was detected to be more suitable.The pixel coordinates of the projectile in the image coordinate system were obtained by using the parameters such as amplitude and amplitude;finally,the passing coordinates of the projectile on the target surface were obtained by coordinate transformation.Compared with other target detection algorithms,this algorithm can effectively improve the detection accuracy of projectiles under the premise of ensuring the detection speed.The detection accuracy rate can reach more than 95%,and has strong applicability,which can be used in various weapons testing.

关 键 词：CCD交汇 目标检测 HOUGH变换 梯度分析 灰度差 

分 类 号：O242.1[理学—计算数学] TP274[理学—数学]