Article ID Journal Published Year Pages File Type
10361716 Pattern Recognition Letters 2005 14 Pages PDF
Abstract
An improved parametric active contour model, FFA snake, is presented in this paper. Based on analyzing the force distribution in the distance potential force field, it is found that there are two types of contour points in the curve. The points in one type are called true contour points and can be classified into six sorts, and the points in the other type are called false contour points. When the energy of the contour is approaching the minimum, the contour is not considered as the final one, and the algorithm must be executed continuously until all points in the curve are judged to be one of the six sorts of the true contour points. Simulation results show that there are three advantages of this model, firstly, it has a large capture range, so it is insensitive to the initialization; secondly, it can move parts of the snake into boundary concavities; finally, it has a low computational cost.
Related Topics
Physical Sciences and Engineering Computer Science Computer Vision and Pattern Recognition
Authors
, ,