A Hopfield-like structure to extract road boundaries from a road image

Abstract In Autonomous Vehicles Problems, a robust method to obtain informations about the position of the road boundaries is required. In this paper, we present a method to classify all the segments of an image based on the possibility that they belong to a road boundary. The term “possibility” is used here with the meaning it has in fuzzy set theory. The classification is accomplished by means of an “iterative relaxation”. The possibility value of each segment is updated according to the possibility values of its neighbor segments.