We have introduced a novel external force model for deform-able models, which we called the gradient vector flow (GVF) field. The field is calculated as a diffusion of the gradient vectors of a gray-level or binary edge map. We have shown that it allows for flexible initialization of the deformable model and encourages convergence to boundary concavities.
Further investigations into the nature and uses of GVF are warranted. In particular, a complete characterization of the capture range of the GVF field would help in deformable model initialization procedures. It would also help to more fully understand the GVF-I parameter ^ and the GVF-II parameter K, perhaps finding a way to choose them optimally for a particular image, and to understand the interplay between the GVF parameters and the deformable model parameters a and ¿6. Also, the GVF framework might be useful in defining new connections between parametric and geometric deformable models, and might form the basis for a new geometric deformable model. Finally, making connections between GVF and other applications in image processing, computer vision, and medical imaging might provide some new insights or even new solutions to existing problems.
Was this article helpful?