单目视觉测量弹目间距及优化方法  

Distance Measurement and Optimization Methods for Projectile and Target Using Monocular Vision

作　　者：许敬[1] 张合[1] 王晓锋[1] 崔业兵[1] 

机构地区：[1]南京理工大学智能弹药技术国防重点学科实验室,江苏南京210094

出　　处：《光学学报》2012年第F12期154-159,共6页Acta Optica Sinica

基　　金：国家自然科学基金（60908037）资助课题.

摘　　要：针对弹丸修正中弹目间距难以测量问题，提出了一种适用于常规弹药的单目视觉测量弹目间距的方法。根据投射成像原理，推算二维目标像点坐标与目标的空间三维位置之间的映射关系。通过分析弹丸、摄像机以及大地坐标系之间的坐标转换关系，建立单目视觉测量弹目间距的数学模型。分析影响距离测量精度的因素，使用二次规划法（SQP）算法优化摄像机焦距，并基于线性变参数（LPV）方法设计非线性滤波器对测量结果进行优化。三轴转台建立弹丸视觉测量的半实物仿真试验表明利用单目视觉测量可以较为准确地提供距离信息，采用LPV方法优化后，能够将误差控制在±2％之内，精度明显提高，从而验证了测量方法的可行性和优化算法的有效性。To solve the problem of measuring the distance between the projectile and the target, a method using monocular vision system for conventional ammunitions to measure the distance is proposed. According to the principle of projection imaging, the mapping relation of the three-dimensional （3-D） position and the 2-D image of the target is derived. The monocular vision measurement mathematical model is established by analyzing the coordinate transformation matrix of the projectile, camera and the reference frame, The affecting factors on the distance measurement accuracy are analyzed. The sequential quadratic programming （SQP） algorithm is used to optimize the focal length of the camera. The nonlinear filter based on linear parameter-varying （LPV） method is adopted to optimize the measurement results. The Semi-physical Simulation measurement experiment is designed on the three- axis turntable. The results show that the monocular vision measurement method can provide the comparatively accurate distance information and the error can be controlled within ± 2 % by using the LPV optimization method which improves the measurement accurate significantly. Then the feasibility of the measurement method and validity of the optimization method are verified.

关 键 词：测量 视觉测量 距离 优化 线性变参数方法 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]