Differential filtering algorithm for robot welding seam image enhancement
Ying Liu, Qingjiu Huang, Kentarō Satō · Journal of Physics Conference Series · 2020
Abstract The low resolution of the welding seam recognition image during robot welding will affect the welding accuracy, and the cost of high-resolution cameras is higher. Therefore, in this paper, the geometrical processing of the bilinear interpolation method for the welded seam pictures taken by the low-resolution camera was used to obtain a high-resolution image. This paper proposed a differential filtering algorithm based on Gaussian filtering and bilateral filtering, to eliminate the fine lines, noise and enhance the weld feature value for the obtained high-resolution image, thereby ensuring the welding accuracy. Finally, through the image processing experiment of increasing the 1.25 million pixel weld image to 5 million pixel weld accuracy, the effectiveness of the image enhancement method proposed in this paper was proved.