Study of technology for detecting seam center based on visual sensor
The experiments were performed to study the technologies of eliminating the noises of seam images obtained from a visual sensor.Some filter methods were analyzed to find the characteristics of the weld image noises.The weld edge coordinates were calculated for those effective filtering weld images.Three image-integrated approaches for detecting the weld characteristics were presented.Also,the efficiency and precision of detecting the seam center of those methods were compared.The experimental results show that the method which adopts Wiener filter and Robert edge detection is the most efficient to obtain the accurate information of the seam center.