Research on Image Processing in Weld Seam Tracking with Laser Structured Light

摘要

According to the image characteristics in weld seam tracking with laser vision, Radon transform is first used to find the location of the laser stripe; then, the transformed image is processed by using the way that the pixel information caused by laser stripe is reserved and the other information by noises is removed, and then inverse Radon transform is applied for image recovery. Finally, improved fast FCM clustering is used for image segmentation so that the groove area can be recognized. These methods can bring good effects with little time cost, and the anti-disturbance performance is well, therefore they are suitable for weld seam tracking with laser vision.