Wednesday, June 23, 2004

Paper Review: Xu98Snakes

Journal = {IEEE Transactions on Image Processing},
Year = {March 1998},
Title = {Snakes, Shapes, and Gradient Vector Flow},
Number = {3},
Pages = {359-369},
Author = {C. Xu and J. L. Prince},
Volume = {7}}
[Abstract] Active contours -- or snakes -- are computer-generated curves that move within images to find object boundaries. We have developed a new kind of snake that permits the snake to start far from the object, and yet still draws it towards the object, and forces it into boundary concavities. The new snake is based on a new type of external force field, called gradient vector flow, or GVF. This field is computed as a spatial diffusion of the gradient of an edge map derived from the image. This computation causes diffuse forces to exist far from the object, and crisp force vectors near the edges. Combining these forces with the usual internal forces yields a powerful computational object: the GVF snake. The 3-D version is called the GVF active surface. We have experimented with GVF on line drawings and grayscale images, including images of the heart and brain. Several examples are shown below and publications and software are also available online.
The former work of this paper is published in a conference paper: .pdf
Journal = {IEEE Proc. Conf. on Comp. Vis. Patt. Recog. (CVPR'97)},
Year = {1997},
Title = {Gradient Vector Flow: A New External Force For Snakes},
Number = {},
Pages = {66-71},
Author = {C. Xu and J. L. Prince},
Volume = {}}

No comments:

Post a Comment