Active Contour Model for the Detection of Sharp Corners in Image Boundaries

Active contours are a form of curves that deform according to an energy minimizing function and are widely used in computer vision and image processing applications to extract features of interests from raw images acquired from an image capturing device. One of the major limitations in an active contour is its inability to converge accurately when the object of interest exhibits sharp corners. In this paper, a new technique to deform the active contour by incorporating prior knowledge of significant corners of the object detected using the Harris operator is introduced. The newly constructed active contour deforms towards the boundary of the object without bypassing sharp corner points. Experimental results with several synthetic images show the ability of the new technique to extract features of interest from images consisting of sharp corners with a high accuracy than traditional methods.