核燃料组件的线结构光主动视觉测量

Line Structured Light Active Vision Measurement for Nuclear Fuel Assembly

根据核燃料上管座尺寸快速、精确检测的要求,对上管座主动视觉检测方法进行了深入研究。依据线激光三角法测量原理,搭建了一套线结构光扫描视觉测量系统。采用Matlab工具箱对相机进行了标定,并对线结构光进行了矫正。采用高斯滤波对图像进行噪声滤除,并改善了阈值提取的过程,避免了单阈值法造成的线结构光提取不完整的缺点。为了进一步提升系统的精度,采用梯度重心法计算出了亚像素坐标位置,结合激光三角法原理计算出待测物体三维点云。根据匹配点计算出旋转平移矩阵完成粗匹配,并使用点云拼接算法进一步提升匹配精度。使用贪婪投影三角形算法完成对点云的三维重建。试验结果表明,该系统能够实时进行测量,且具有较高的测量精度;位于0. 5 m处求解出的燃料组件特征尺寸平均误差小于0. 5 mm,满足检测要求。

According to the requirement of fast and accurate measurement of the sizes of the upper tube seat of nuclear fuel,the active visual detection method of the upper tube seat is studied in depth.In accordance with the principle of line laser triangulation,a line structured light scanning visual measurement system is built.The camera is calibrated by Matlab toolbox,and the line structured light is corrected.Gauss filter is used to remove noise from the image,and the process of threshold extraction is improved,which avoids the defect of incomplete extraction of line structured light caused by single threshold method.In order to further improve the accuracy of the system, the gradient barycenter method is used to calculate the subpixel coordinate position,and the laser triangulation method is used to calculate the three-dimensional point cloud of the object to be measured.According to the matching points,the rotation translation matrix is calculated to complete rough matching,and the point cloud stitching algorithm is used to further improve the matching accuracy.The greedy projection triangle algorithm is used for 3D reconstruction of the point cloud.The experimental results show that the system can measure in real time and has high measurement accuracy.The average error of the characteristic sizes of the fuel assembly calculated at 0.5 m is less than 0.5 mm,which meets the detection requirement.