Harris function based active contour external force for image segmentation
Deformable active contour (snake) models are efficient tools for object boundary detection. Existing alterations
of the traditional gradient vector flow (GVF) model have reduced sensitivity to noise, parameters
and initial location, but high curvatures and noisy, weakly contrasted boundaries cause difficulties for
them [Fig. 1(b)].
Two Harris based parametric snake models were introduced, Harris based gradient vector flow (HGVF) and Harris based vector field convolution (HVFC), which use the modified version of the curvature-sensitive Harris characteristic function [Fig. 1(c)] to achieve a balanced, twin-functionality (corner and edge) feature map [Fig. 1(d)]. To avoid initial location sensitivity, starting contour is defined as the convex hull of the most attractive points of the map [Figs. 1(e) and 1(f)].
Results show that our methods outperform the classical approaches, when tested on images with high curvature, noisy boundaries [Fig. 2].
A. Kovacs and T. Sziranyi, "Harris Function Based Active Contour External Force for Image Segmentation", Pattern Recognition Letters, vol. 33, no. 9, pp. 1180-1187, 2012, IF:1.266
Email: andrea.manno-kovacs AT sztaki.mta.hu