基于改进Steger算法的Ⅴ型焊缝识别方法  

V-Shaped Weld Identification Method Based on Improved Steger Algorithm

在线阅读下载全文

作  者:郭北涛[1] 刘磊 张丽秀[2] 刘瀚齐 金福鑫 GUO Beitao;LIU Lei;ZHANG Lixiu;LIU Hanqi;JIN Fuxin(School of Mechanical and Power Engineering,Shenyang University of Chemical Technology,Shenyang 110142,China;School of Transportation and Mechanical Engineering,Shenyang Jianzhu University,Shenyang 110168,China)

机构地区:[1]沈阳化工大学机械与动力工程学院,沈阳110142 [2]沈阳建筑大学交通与机械工程学院,沈阳110168

出  处:《组合机床与自动化加工技术》2025年第3期182-186,共5页Modular Machine Tool & Automatic Manufacturing Technique

基  金:国家自然科学基金项目(51375317);沈阳市科技计划项目(F16-228-6-00)。

摘  要:针对采用线结构光进行Ⅴ型焊缝特征识别时准确性不足和效率低的问题,提出了一种基于改进Steger算法的Ⅴ型焊缝识别方法。通过改进Steger算法流程和自适应感兴趣区域的选择,提升了Ⅴ型焊缝特征点提取精度和效率。首先,通过对焊缝图像进行平滑处理和阈值分割操作,将结构光条纹与背景信息分离;其次,采用一种新的感兴趣区域选择方法,有效地找出了结构光条纹所在区域;然后,采用骨架细化法和Canny边缘检测算法估算出光条纹宽度,根据宽度信息分割光条纹求取光条中心线;最后,采用最小二乘法得到Ⅴ型焊缝特征点的亚像素坐标。实验结果表明,改进Steger算法运行速度提升约60%,平均均方根误差降低0.23 pixels,提高了Steger算法的速度和精度。Ⅴ型焊缝识别误差均小于1.0 mm,证明了该方法能够准确地识别出Ⅴ型焊缝特征,满足实际焊接精度要求。In order to solve the problems of low accuracy and low efficiency in V-shaped weld feature recognition by using linear structured light,a V-shaped weld recognition method based on improved Steger algorithm was proposed.By improving the process of the Steger algorithm and the selection of the adaptive region of interest,the extraction accuracy and speed of the Steger algorithm are improved,and the accuracy and efficiency of the extraction of V-shaped weld feature points are improved.Firstly,the noise information was filtered and the structured light fringes were separated from the background information by smoothing and threshold segmentation of the weld image.Secondly,a new region of interest selection method is adopted to effectively find out the region where the structured light fringes are located,which reduces the waste of computation and improves the image processing efficiency.Then,the skeleton refinement method and the Canny edge detection algorithm were used to estimate the width of the light stripes,and the light stripes were segmented according to the width information to obtain the center line of the light stripes.Finally,the intersection point is calculated by fitting the straight line by the least squares method,and the sub-pixel coordinates of the V-shaped weld feature point are obtained.Experimental results show that the running speed of the improved Steger algorithm is increased by about 60%,and the average root-mean-square error is reduced by 0.23 pixels,which improves the speed and accuracy of the Steger algorithm.The identification errors of V-shaped welds are all less than 1.0 mm,which proves that the method can accurately identify the characteristics of V-shaped welds and meet the requirements of actual welding accuracy.

关 键 词:线结构光 Steger 感兴趣区域 直线拟合 特征提取 

分 类 号:TH162[机械工程—机械制造及自动化] TG506[金属学及工艺—金属切削加工及机床]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象