基于激光双目视觉的接缝三维重建

Binocular stereo vision with laser to reconstruct welding seam

关键词:  接缝重建  双目视觉  三维重建  图像处理  结构激光视觉

Key words:  reconstruction of seam  binocular stereo vision  3D reconstruction  image processing  structure laser vision

摘要:

设计研制了结构激光双目视觉传感器,该传感器加入结构激光,很好地解决了立体视觉中对应点的匹配问题。利用该传感器采集接缝图像,开发了图像采集处理程序,实现接缝图像处理和特征分析,识别并提取了二维平面上接缝特征点的图像坐标。采用MATLAB软件实现了三维重建标定算法,根据双目视觉原理,由接缝特征点的图像坐标计算出其三维坐标,进行了三维重建,抽取了接缝三维几何尺寸、坡口类型、角度、间隙等焊接特征信息。

Abstract:

A binocular stereo vision with laser system was developed.Clear image of the welding seam was obtained,The difficult problem of the Feature Point Matching algorithm in stereo vision was solved with the structure laser.A programme to acquire and process image was developed by Visual C++.The 2D image coordinates of the seam feature and the character of the seam was obtained.An algorithm of 3D coordinates calculation method was proposed.It can reconstruct the 3D coordinates of the seam with the 2D image coordinates of the seam feature point and obtain precise information of the seam.The experiments of the reconstruction of typical seams were done.The results show that the accuracy of the seam reconstruction is very high.