基于自适应双边滤波的V型焊缝特征提取研究

Research on feature extraction of V-shaped welds based on adaptive bilateral filter

  • 摘要: 为了提高自动化焊接时焊缝特征提取的精准度,提出一种自适应双边滤波算法对图像噪声去除。首先在图像预处理阶段使用自适应双边滤波算法对采集到的焊缝原始图像进行去噪处理,然后通过阈值分割得到激光条纹的二值化图像,将灰度重心法与最小二乘法相结合提取出激光条纹中心线,最后通过联立激光条纹中心线方程组求解焊缝角点坐标,从而得到焊接点位置坐标。实验结果表明,该方法能够准确地提取出V型焊缝的特征角点坐标,精度上满足实际的焊接要求。

     

    Abstract: In order to improve the accuracy of welding seam feature extraction during automatic welding, an adaptive bilateral filtering algorithm is proposed to remove image noise. Firstly, use the adaptive bilateral filter algorithm to denoise the original image, and the binary image of laser stripe is obtained by threshold segmentation. The center line of laser stripe is extracted by combining gray centroid method and least square method. Finally, the coordinates of welding point are obtained by solving the equations of center line of laser stripe. The experimental results show that the method can accurately extract the characteristic corner coordinates of the V-shaped weld, and the accuracy can meet the actual welding requirements.

     

/

返回文章
返回