A syntactic and contextual edge detector and application of a rewriting rules set to improve detection

This paper presents an image segmentation method which may be applied to various tasks as natural segmentation of monochromatic, colored and even three dimensional images. This algorithm is based on syntactic recognition. The particular novelty of this algorithm lies in the use of an efficient grammar to find edge models with an unknown a priori context. In order to improve and eventually to correct the detection, a set of rules is applied on the edges image. Experimental results are shown.<
>