基于惯性椭圆拟合的小目标姿态测量方法  

Attitude Measurement Method Based on Inertia Ellipse Fitting for the Small Targets

作　　者：金国强 陈宇璇 阚艳 金一[1] 竺长安[1] JIN Guoqiang;CHEN Yuxuan;KAN Yan;JIN Yi;ZHU Chang’an(School of Engineering Science,University of Science and Technology of China,Hefei 230026,China;State-Owned Wuhu Machinery Factory,Wuhu,Anhui 241000,China)

机构地区：[1]中国科学技术大学工程科学学院,合肥230026 [2]国营芜湖机械厂,安徽芜湖241000

出　　处：《计算机工程》2021年第4期211-217,225,共8页Computer Engineering

基　　金：国家重大科研仪器研制项目(61727809)。

摘　　要：视觉测量方法在飞行器姿态测量中被广泛应用,然而远距离飞行目标成像纹理缺失会使传统光测手段应用受限。对测量中采用的追踪手段、图像分割技术以及测量方法进行研究,提出一种改进的小目标姿态测量方法。利用高速相机拍摄目标,通过跟踪算法对运动目标进行框选,使用非面积极大区域抑制法分割目标,计算目标图像的惯性椭圆从而实现目标姿态测量。实验结果表明,在仿真情况、低复杂度及高复杂度的真实拍摄场景下,该方法的误差分别低于0.233°、0.307°及0.325°,基本满足导弹等小目标的姿态测量精度需求,测量结果稳定可靠,能够有效降低人工判读的工作量。The visual measurement methods are widely used in aircraft attitude measurement.However,the performance of traditional optical measurement methods is limited due to the lack of texture information of longdistance flying targets.To address the problem,this paper proposes an improved small target attitude measurement method.The method uses the high-speed camera to capture the target,and then the target tracking algorithm to locate the moving targets.Next the non-area maximal region suppression method is used to segment the target,and the inertial ellipse of the target image is calculated to measure the attitude of the target.The experimental results show that the error of the proposed method is within 0.233°in the simulated shooting scene,0.307°in the actual shooting scene of low complexity,and 0.325°in the actual shooting scene of high complexity.The method satisfies the basic accuracy requirements of attitude measurement of small targets such as missiles,and its measurement results are stable and reliable enough to reduce the manual interpretation workloads significantly.

关 键 词：目标姿态测量 小目标 惯性椭圆 光学测量 高速摄像 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]