不同分布下红外主动目标特征点配准方法研究  

Research on registration method of infrared active target feature points under different distributions

作　　者：范帅鑫 谷玉海[1] 邹志[2] 崔悦 Fan Shuaixin;Gu Yuhai;Zou Zhi;Cui Yue(Key Laboratory of Modern Measurement and Control Ministry of Education,Beijing Information Science and Technology University,Beijing 100192,China;Changcheng Institute of Metrology&Measurement,Beijing 100095,China)

机构地区：[1]北京信息科技大学现代测控技术教育部重点实验室,北京100192 [2]航空工业北京长城计量测试技术研究所,北京100095

出　　处：《电子测量技术》2024年第6期71-78,共8页Electronic Measurement Technology

基　　金：工业和信息化部民用飞机专项科研技术研究项目(MJ-2018-J-70);北京市科技委促进高校内涵发展-学科建设专项资助项目(5112011015);机电测控系统北京市重点实验室开放课题(KF20202223204)项目资助。

摘　　要：视觉测量中经常需要用到特征匹配来计算位姿信息,但尚无针对红外主动目标设计特征匹配可用的算法,为实现对不同分布红外主动目标的匹配,本文提出了一种通用的两阶段特征点匹配方法。第一阶段为粗配准,首先检测图像特征点集的凸包,获取最外围点;通过构建三角形特征集,并使用马氏距离进行相似三角形计算搜索实现快速粗配准。第二阶段为精匹配,首先通过粗匹配特征计算欧拉角避免匹配结果的180°旋转对称;针对粗配准后的可能存在的特征点缺失问题,采用极线约束精匹配策略,充分利用已匹配特征点的几何信息,有效实现对剩余点的精确匹配。理论分析与实验表明,在13个红外发光点组成的旋转对称点集及非旋转对称点集下,该方法在绝对大小0°~40°的旋转范围内能高效匹配,实验测试极限性能能够达到50°,并对实际场景下特征点的遮挡等情况具有较好的鲁棒性,实验结果验证其适应性与稳定性,具有较高的实用价值。Feature matching is often used to calculate pose information in visual measurement,but there is no available algorithm for designing feature matching for infrared active targets.In order to achieve matching of infrared active targets with different distributions,this paper proposes a general two-stage feature Point matching method.The first stage is coarse registration.First,the convex hull of the image feature point set is detected to obtain the outermost points.Fast coarse registration is achieved by constructing a triangle feature set and using Mahalanobis distance to calculate and search for similar triangles.The second stage is precise matching.First,the Euler angle is calculated through coarse matching features to avoid the 180°rotational symmetry of the matching results.In order to solve the problem of possible missing feature points after coarse registration,the epipolar constraint fine matching strategy is adopted to make full use of the existing features.Match the geometric information of feature points to effectively achieve accurate matching of remaining points.Theoretical analysis and experiments show that under the rotational symmetry point set and the non-rotation symmetry point set composed of 13 infrared luminescent points,this method can efficiently match within the absolute rotation range of 0°~40°,and the experimental test limit performance can reach 50°,and has good robustness to the occlusion of feature points in actual scenes.The experimental results verify its adaptability and stability,and has high practical value.

关 键 词：红外主动目标 点集匹配 仿射变换 旋转对称 对极约束 遮挡点 

分 类 号：TH741[机械工程—光学工程]