Segmentation of lines using perceptual organization with active contour functions

Abstract

We present a new method for perceptual grouping of lines using their geometric properties. First the visual properties expected from the groupings are modelled as a quality function similar to active contour functions. This function is computed recursively and optimized from a local to global level with an algorithm related to dynamic programming. The final groupings are then selected according to their global quality. Applied to various situations, the method proved its adaptability and its robustness to noisy environments. The results showed how the use of visual properties can provide an effective segmentation before a high level image understanding.