Research on Watershed Algorithm Based on Image Marking Method Optimization
Haibo Lin, Shuang Song, Shanli Tao, Haoran Liu · 2021
In image processing, due to the presence of noise or other unpredictable interference factors, the use of traditional watershed algorithms often results in excessive segmentation. In order to solve this problem, the watershed algorithm optimized based on the image tagging method can be used to guide the computer to complete the watershed algorithm segmentation by artificially setting the prior knowledge in order to obtain a better image segmentation effect. Finally, it is verified by experiments that the watershed algorithm based on the image labeling method is better than the traditional watershed algorithm in segmenting some complex images.