Active Contour Model

The active contour model is a method to fit closed splines, or snakes, to lines or edges in an image that has been widely applied to image segmentation tasks and analysis. Snakes are energy minimizing, deformable splines influenced both by image forces that pull it towards object contours and by constraint, internal forces that resist deformation. Snakes may be understood as a special case of the general technique of matching a deformable model to an image by means of energy minimization. However, snakes do not solve the entire problem of finding contours in images, since the method requires knowledge of the desired contour shape beforehand. Rather, they depend on other mechanisms. In Dragonfly, snakes must be initiated to the approximate shape of the boundary and then iterated to pull it towards the object contours.

In the example below, a series of paths that describe the femoral head, neck, and top of the thigh bone was first added to the 2D views of a pelvic CT scan and then a surface mesh was generated from the selected paths.

Active contour mesh (in blue)

Compared to classical feature extraction techniques, snakes have multiple advantages:

• They adaptively search for the minimum state.
• External image forces act upon the snake in an intuitive manner.
• Incorporating Gaussian smoothing in the image energy function introduces scale sensitivity.

The key drawbacks of snakes are:

• They are sensitive to local minima states, which can be counteracted by simulated annealing techniques.
• Minute features are often ignored during energy minimization over the entire contour.
• Their accuracy depends on the convergence policy.

[1] Michael Kass, Andrew Witkin, Demetri Terzopoulos, Snakes: Active contour models, International Journal of Computer Vision, 1, 4, (1988).

[2] Active Contour Model (https://en.wikipedia.org/wiki/Active_contour_model)