Improved OTSU method on welding seam image segmentation

Image segmentation is important in comprehension and analysis of image.Threshold segmentation is widely used in image segmentation because of its simplicity and high efficiency.How to get the fine threshold is the key of image threshold segmentation.The OTSU algorithm(Maximization of interclass variance)is one of the superior threshold selection methods.When the OTSU method is used for welding seam image segmentation,the quality of image segmentation is not good because the welding seam images often appear gray level asymmetry of large area.Based on welding seam image analysis and OTSU algorithm,an improved method of adaptive threshold selection is proposed.The experiment results show that the new method is strongly adaptable and efficient for welding seam image segmentation.