3D Watershed Based on Rainfall-Simulation for Volume Segmentation

This paper proposes a 3D watershed algorithm based on the 2D rainfall-simulation based watershed. It shares the same procedure of finding the steepest descending path, with the difference being the number of connectivity and the location of the neighbours. The aim of the 3D watershed is to segment out volumes from a 3D space. This kind of segmentation can be useful in areas such as medical imaging where the segmentation of the different volumes from a scan can aid doctors in patient diagnosis, advanced surgical planning and simulation, radiotherapy planning, etc.