基于改进贝塞尔曲线的弯管视觉识别方法研究
首发时间:2019-02-20
摘要:为了解决弯管曲率半径一致性较差引起的自动化抛光中质量不稳定现象,提出了一种改进贝塞尔曲线拟合弯管中线的新方法,解决了弯管自动化抛光中视觉识别的关键问题。在通过边缘角点定位算法定位弯管角点的基础上,根据角点位置分离弯管内外轮廓并分别统计内外轮廓位置点,提出弯管中线定位方法,利用内外轮廓点计算中线坐标并进行坐标转换,最后采用最小二乘法及改进贝塞尔曲线对弯管的中线进行分段拟合,经与三坐标测量结果比较,曲线拟合的精度满足砂带抛光要求。
For information in English, please click here
Research on Visual Recognition Method of Bent Pipe Based on Improved Bezier Curve
Abstract:In order to solve the problem of quality instability in automatic polishing caused by poor consistency of curvature radius of curved pipe, a new method for fitting the midline of bent pipe through improving Bezier curve is proposed, which solves the key issue of visual recognition in automatic polishing of bent pipe. On the basis of positioning the corner point of the bent pipe by the edge corner location algorithm, the inner and outer contours of the bent pipe are separated according to the position of the corner point, and the position points of the inner and outer contours are respectively counted. The method of positioning the center line of the bent pipe is proposed. The coordinates of the center line are calculated by using the inner and outer contour points and the coordinates are transformed. At last, the least square method and the modified Bezier curve are used to segmentally fit the midline of the bent pipe. Compared with the three-coordinate measurement results, the accuracy of the curve fitting satisfies the requirements of abrasive belt polishing.
Keywords: edge corners elbow midline Bezier curve midline fitting
基金:
引用
No.****
动态公开评议
共计0人参与
勘误表
基于改进贝塞尔曲线的弯管视觉识别方法研究
评论
全部评论0/1000