基于斜率单调区间跨度分割的焊缝轮廓特征点的提取
|
摘要
焊缝轮廓特征点是厚板机器人焊接中指导焊枪实施焊道规划和焊缝跟踪的依据。以提取的T形接头焊缝轮廓为研究对象,提出了一种利用斜率单调区间跨度Otsu分割来提取焊缝轮廓特征点的算法。该算法首先利用有效的斜率计算方法求取预处理后的焊缝轮廓的斜率,然后对斜率进行平滑处理并确定斜率单调区间,从中筛选出斜率跨度超过平均跨度的区间;将这些区间的跨度再进行Otsu分割以确定跨度超过获得的阈值区间,获得最终区间的中间位置即为焊缝轮廓的特征点。该算法无需设定任何参数,可用于典型的焊缝轮廓特征点的提取。
Feature points of the weld profile are the guide for bead planning and seam tracking in robotic welding of thick steel plates. In this paper,the extracted weld profile of T-joints was studied,and a method to extract feature points of the weld profile of T-joints was proposed based on span segmentation of slope monotone intervals using Otsu. Firstly,slopes are calculated effectively using the preprocessed weld profile data,and followed by smooth processing. Then these intervals whose slope spans are bigger than the average one are selected. In order to determine the final slope monotone intervals from the selected ones,the slope span values are segmented by Otsu,the centers of the intervals can be deemed as the feature points of the weld profile. The advantages of the method are that it can run well without any empirical parameters and can be applied to the typical welded joints.
Feature points of the weld profile are the guide for bead planning and seam tracking in robotic welding of thick steel plates. In this paper,the extracted weld profile of T-joints was studied,and a method to extract feature points of the weld profile of T-joints was proposed based on span segmentation of slope monotone intervals using Otsu. Firstly,slopes are calculated effectively using the preprocessed weld profile data,and followed by smooth processing. Then these intervals whose slope spans are bigger than the average one are selected. In order to determine the final slope monotone intervals from the selected ones,the slope span values are segmented by Otsu,the centers of the intervals can be deemed as the feature points of the weld profile. The advantages of the method are that it can run well without any empirical parameters and can be applied to the typical welded joints.
引文
[1]Chen H,Liu W,Huang L,et al.The decoupling visual feature extraction of dynamic three-dimensional V-type seam for gantry welding robot[J].The International Journal of Advanced Manufacturing Technology,2015,80(9-12):1741-1749.
[2]黎咸西,熊震宇.基于视觉传感的多层多道焊缝图像特征的识别[J].南昌航空大学学报(自然科学版),2011,25(1):53-57.
[3]檀朝彬,曾惠林,皮亚东,等.基于激光结构光的管道激光/电弧复合焊焊道跟踪与能量控制系统[J].焊接,2015(7):26-29.
[4]刘国平,张华,胡2)华,等.V形焊缝特征点提取的算法研究[J].焊接,2006(2):36-38.
[5]Zaojun F,De X,Min T.A vision-based self-tuning fuzzy controller for fillet weld seam tracking[J].IEEE/ASME Transactions on Mechatronics,2011,16(3):540-550.
[6]雷正龙,吕涛,陈彦宾,等.基于扫描激光视觉传感的焊缝图像特征信息识别[J].焊接学报,2013,34(5):54-58.
[7]刘航,李志勇,任杰亮,等.焊缝轮廓线激光检测系统开发及算法实现[J].焊接,2017(1):27-31.
[2]黎咸西,熊震宇.基于视觉传感的多层多道焊缝图像特征的识别[J].南昌航空大学学报(自然科学版),2011,25(1):53-57.
[3]檀朝彬,曾惠林,皮亚东,等.基于激光结构光的管道激光/电弧复合焊焊道跟踪与能量控制系统[J].焊接,2015(7):26-29.
[4]刘国平,张华,胡2)华,等.V形焊缝特征点提取的算法研究[J].焊接,2006(2):36-38.
[5]Zaojun F,De X,Min T.A vision-based self-tuning fuzzy controller for fillet weld seam tracking[J].IEEE/ASME Transactions on Mechatronics,2011,16(3):540-550.
[6]雷正龙,吕涛,陈彦宾,等.基于扫描激光视觉传感的焊缝图像特征信息识别[J].焊接学报,2013,34(5):54-58.
[7]刘航,李志勇,任杰亮,等.焊缝轮廓线激光检测系统开发及算法实现[J].焊接,2017(1):27-31.