基于双目视觉的正交函数局部拟合的工件定位方法

工件定位是自动化生产线的重要过程,是后续分拣、装配等操作的基础。为了解决传统的工件定位误差较大的问题,提出一种基于双目视觉的正交函数局部拟合的工件定位方法。系统建立之前,采集一张拟合板的图像,获得位置信息作为拟合的样本点。工件的定位首先要采集待定位的工件图像,进行图像预处理,通过特征提取获得工件形心的像素值,然后利用样本点进行正交函数局部最小二乘拟合,获得工件形心的平面坐标,最后通过双目重构,获得工件形心的空间坐标。实验结果表明:基于双目视觉的正交函数局部拟合的工件定位方法在精度上比传统的相机标定、三维重建的定位方法有明显提高。该方法不仅提高了工件定位精度,而且避免了标定环节,具有较好的应用价值。

Workpiece positioning method of local fitting of orthogonal function based on binocular vision

Aiming at the workpiece positioning in the automatic production line, a method of local fitting of orthogonal function based on binocular vision was proposed. Before the workpiece positioning, an image of a fitting plate was acquisited and processed, and the position information was obtained as the sample points. Firstly, the image of the workpiece was collected and preprocessed, and the pixel centroids of the workpiece were obtained by feature extraction.Then the plane coordinates were obtained by local fitting of orthogonal function.Finally, the space coordinates of the workpiece were obtained by binocular reconstruction. The experimental results show that the proposed fitting method is more accurate than the traditional method by camera calibration and 3D reconstruction, and avoids the calibration. It has good practical application value.