摘要
利用机器视觉进行飞机蒙皮损伤检测是提高其自动化水平的有效手段,其中飞机蒙皮全局三维重建是检测过程的关键步骤,可定位损伤部位的精确位置。为解决当前技术所存在的设备复杂、拼接精度低、处理效率低等问题,提出一种基于后方定位并结合结构光对飞机蒙皮进行快速三维重建的方法。通过结构光三维测量系统对飞机蒙皮局部区域进行三维重建,采用后方摄像机同步观测结构光系统,并对其空间位姿进行定位。在空间位姿数据的辅助下,将结构光系统每次测量的飞机三维形貌数据融合至定位摄像机坐标系下,实现对大飞机蒙皮三维形貌的快速、高精度非接触式测量,这为飞机蒙皮损伤的自动化视觉检测提供了有效的技术支持。
It is an effective method to improve the automation level of aircraft skin damage detection by using machine vision.The global three-dimensional(3D)reconstruction of aircraft skins is the key step in the detection process to locate the exact location of the damaged parts.In order to solve the problems of complex equipment,low splicing accuracy,and low processing efficiency in current technology,this paper proposes a rapid 3D reconstruction method for aircraft skins based on rear positioning and combined with structured light.The 3D reconstruction of the local area of aircraft skins is conducted by using the structured light 3D measurement system,and the rear camera synchronously observes the structured light system to determine its spatial position and posture.With the help of spatial pose data,the 3D topography data of the aircraft measured by the structured light system is fused into the positioning camera coordinate system to realize the rapid and high-precision non-contact measurement of the 3D topography of the large aircraft skin.This method provides effective technical support for the automatic visual inspection of aircraft skin damage.
作者
吴军
李鑫
刘少禹
李雁玲
于之靖
Wu Jun;Li Xin;Liu Shaoyu;Li Yanling;Yu Zhijing(Aeronautical Engineering College,Civil Aviation University of China,Tianjin 300300,China;State Key Laboratory of Precision Measuring Technology and Instruments,Tianjin University,Tianjin 300072,China;Electronic Information and Automation College,Civil Aviation University of China,Tianjin 300300,China)
出处
《光学学报》
EI
CAS
CSCD
北大核心
2021年第11期123-131,共9页
Acta Optica Sinica
基金
国家自然科学基金(52005500)
中央高校科研基本业务费(3122019088)
天津市教委科研计划项目(2018KJ242)。
关键词
机器视觉
三维重建
视觉定位
数据融合
蒙皮损伤检测
machine vision
three-dimensional reconstruction
visual positioning
data fusion
skin damage detection
