CAJ | 학술논문

焊缝图像的边缘提取是实现焊缝视觉跟踪的一个重要环节.用抗噪膨胀腐蚀型形态学算子对单边V型坡口横焊焊缝二值图像进行了边缘提取,并与传统的Roberts算子、Sobel算子、Canny算子、LOG算予以及典型的形态学腐蚀算子、膨胀算子和膨胀腐蚀算子的边缘提取效果进行了对比.结果表明,抗噪膨胀腐蚀型算子具有非常强的抗噪性,可以有效地提取出焊缝图像边缘,更利于后续的焊缝特征点提取.
한봉도상적변연제취시실현한봉시각근종적일개중요배절.용항조팽창부식형형태학산자대단변V형파구횡한한봉이치도상진행료변연제취,병여전통적Roberts산자、Sobel산자、Canny산자、LOG산여이급전형적형태학부식산자、팽창산자화팽창부식산자적변연제취효과진행료대비.결과표명,항조팽창부식형산자구유비상강적항조성,가이유효지제취출한봉도상변연,경리우후속적한봉특정점제취.
The edge detection of seam image is an important step to realize the vision seam tracking. The edge detection of seam binary image in horizontal position welding of single bevel groove was implemented by using an improved morphological edge detection operator:anti-noise-dilation-erosion operator. The edge detection result was compared with those of traditional detection operators, such as Roberts operator, Sobel operator, Canny operator, LOG operator etc, and those of typical morphological edge detection operators,such as erosion operator,dilation operator and dilation-erosion operator,respectively. The results show that the anti-noise-dilation-erosion operator has a good anti-noise property,and can detect the edge of seam image effectively ,which is beneficial to the extraction subsequent image features.