An Improved Algorithm of Contour Initialization in Active Contour Model

Abstract

An improved algorithm of Contour Initialization in GVF Snake Model is proposed in this paper based on detailed analysis of the problems in current active contour models: poor performance under interference, large computational cost, inability to converge to complex or rough contours and difficulty for setting initial contour, etc. Our method efficiently solved the problems in complex or rough contours initialization by connecting the initial points of different resolutions obtained by multi-scale edge detection methods based on wavelet transform and suitably controlling the number of GVF force field iterations. Experiments show that the initial contour can be accurately set close to the objective contour and successfully extract the objective contours with interferential edges and points. The new contour initialization algorithm has further reduced computational cost, improved the anti-jamming ability and extended the application range of GVF Snake model.