基于改进结构光条纹中心提取的三维视觉检测技术

3D Vision Inspection Technology Based on Improved Structured Light Stripe Centre Extraction

针对工业金属零部件表面质量检测中采集的激光条纹图像存在灰度分布不均等问题,提出一种基于线结构水平扫描结合改进激光条纹中心提取的三维检测技术。基于调制的线结构光三维计算机视觉技术,标定工业相机、线结构光平面及水平扫描连续帧的位姿变换;基于窗口的二维动态阈值灰度加权改进激光条纹中心线提取方法,提高中心线提取精度。实验结果表明:采用改进结构光条纹中心提取的三维视觉检测技术重建光滑陶瓷地板表面点云,深度标准差为0.09 mm;采用该技术对高度10 mm的金属标准块表面形貌进行检测,检测精度优于0.3 mm,误差在合理范围内。提出的三维检测技术可高效无损伤地实现高精度工业金属零部件的三维检测。

Aiming at the problem of uneven gray distribution of laser stripe images collected in the surface quality inspection of industrial metal parts,a 3D inspection technology based on line structure horizontal scanning combined with improved laser stripe center extraction was proposed.The 3D computer vision technology based on modulated linear structured light was used to calibrate the position and attitude transformation of industrial cameras,the linear structured light planes and horizontal scanning continuous frames.The method of laser stripe centerline extraction was improved by using 2D dynamic threshold gray scale weighting based on window,and the accuracy of centerline extraction was improved.The experimental results show that the 3D vision inspection technology with improved structured light stripe centreline extraction is used to reconstruct the point cloud of smooth ceramic floor surface,and the standard deviation of depth is 0.09 mm.This technology is used to detect the surface morphology of metal standard blocks with a height of 10 mm,and the detection accuracy is better than 0.3 mm,and the error is within a reasonable range.The proposed 3D inspection technology can efficiently and undamaged realize the threedimensional detection of high-precision industrial metal parts.