论文部分内容阅读
针对平面零件在背光源照射下成像存在伪边缘而造成测量精度下降的问题,提出了一种从被测物伪边缘中恢复准确边缘的测量方法.首先,采用Canny算子提取平面零件的像素边缘,在此基础上利用高斯曲面拟合算法实现亚像素边缘定位,得到伪边缘;然后,提出一种基于方向角差值的边缘点判别法从伪边缘中区分上下边缘点并提出欧氏距离平均值法获得被测物上边缘;最后,利用相似三角形法将上边缘转换到测量平面,结合摄像机标定参数求解平面零件的尺寸.实验结果表明,在1200 mm×800 mm的视野范围内,对不同厚度的平面零件进行尺寸测量,绝对误差小于0.08 mm,相对误差小于0.05%.“,”Aiming at the problem that the imaging accuracy of the planar part was decreased due to pseudo edges in the illumination of the backlight, a measurement method for recovering the accurate edge from the pseudo edges of the measured object was proposed. Firstly, pseudo sub-pixel edges of planar parts were extracted by Canny algorithm and surface fitting algorithm. Secondly, upper and lower edge points distinguishing method based on direction angle difference was proposed and the upper edges were restored by the method of mean Euclidean distance. Finally, the upper edges were converted to the measurement plane by the similar triangular method, and dimension of planar parts could be calculated by using the camera calibration parameters. The results show that the absolute error is less than 0.08 mm and the relative error is less than 0.05% in the range of 1 200 mm×800 mm.