基于图像处理的野外金属破片靶测量研究  

Research on Field Metal Fragment Target MeasurementBased on Image Processing

作　　者：何淼 王浩[1] 黄勇[1] 薛绍[1] HE Miao;WANG Hao;HUANG Yong;XUE Shao(School of Energy and Power Engineering,Nanjing University of Science and Technology,Nanjing 210094,China;Institute of Artificial Intelligence,Nanjing Vocational College of Information Technology,Nanjing 210043,China)

机构地区：[1]南京理工大学能源与动力工程学院,江苏南京210094 [2]南京信息职业技术学院人工智能学院,江苏南京210043

出　　处：《弹道学报》2021年第4期83-90,共8页Journal of Ballistics

基　　金：国防基础加强项目;江苏省2021年青蓝工程中青年学术带头人培养计划。

摘　　要：为解决野外破片战斗部爆破试验中大型靶板毁伤面积人工测量低效且工作量巨大的问题,提出基于逆透视变换图像矫正、均值-中值组合滤波的2D-Otsu鲁棒阈值分割、膨胀腐蚀开闭运算、邻域搜索生长法等图像处理技术,进行靶板毁伤区识别,实现了破片靶图高精度面积测量,克服了野外靶图取景畸变、锈蚀、金属纹理等影响测量精度问题。该方法针对野外金属破片靶的孔洞数量统计及面积测量具有稳定易用易部署的优势。In order to solve the problems of low efficiency and huge workload for manually measuring the damage area of large target plates during burst experiments by filed fragment combat units,the imaging processing technologies such as image correction based on inverse perspective transformation,2D-Otsu robust threshold segmentation based on mean-median combined filter,expansion corrosion opening and closing operation,neighborhood searching and growing method,were applied to identify the target demage area.High precision area measurement of fragment target map was realized.It overcomes the problems affecting the measurement accuracy,such as viewfinder distortion,corrosion,metal texture and so on.The proposed method is targeted towards the design of field fragmented metal target.It is stable,and it is easy to use and deploy.

关 键 词：毁伤评估 图像矫正 抗锈分割 邻域搜索生长法 靶板识别 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]