Abstract: For the real image with strong noise, the cases often appear that there are branches on the useful lines and that noise line consists of many branches in the thinned image. It is difficult to extract accurate information from the useful line with branches and to remove the noise line comprising many branches via size filter. Aiming at solving these problems, the strategy that is firstly to break the branches and then to remove them by using size filter is taken, and a breaking branches algorithm is proposed. The function of the algorithm is to break the lateral branch from the two main branches. The results of welding seam image processing show that the branches can be broken quickly and effectively by using this algorithm, which is helpful to the stability of the continuous image processing.

Key words: Breaking branches algorithm, Image processing, Lateral branch line, Thinned image