Improvement on curve evolution method based on edgeflow
Jing Wang · Journal of Shaanxi University of Technology · 2008
Edgeflow-driven curve evolution is one of the efficient models for image segmentation.This model mainly depends on the smoothing effect of Gaussian function to create the edgeflow vector field.Here we calculate the edgeflow vector field by using cubic B-spline function.Unlike the non-compact support characteristic of Gaussian function,the cubic B-spline function is a compact support and the property makes them useful for computer to deal with.Experiment on computer-produced images and real images shows that the improved model is very efficient.