Quality inspection algorithm based on machine vision for tube-sheet welding

In order to realize the on-line quality inspection of tube-sheet welding, this paper establishes a vision-based quality inspection system to realize the defect detection. And image processing, as the most important part of inspection system, is in-depth studied. Image processing algorithms include two parts: preprocessing and defect detection. Image preprocessing includes median filter, Otsu segmentation, morphological operation, ROI (region of interest) and area smooth. In the defect detection part, the sum intensity of tube is used to judge whether there is a defect and the position of defect is determined accurately. Experimental results show that the image processing method can be widely used with high reliability and robustness.