An optimal method for automatic road updating from high resolution satellite image

This paper proposes an optimal mathematical model for updating road sides from high resolution satellite images. Based on this model, The Ant Colony Optimization (ACO) approach is first used to establish a pheromone matrix that represents the edge information at each pixel position of the image. Next the line plausibility map is used as the external energy in a snake model to identify and extract roads sides of image. Snake is a parametric curve which is allowed to deform from some arbitrary initial location toward the desired final location by minimizing an energy function based on the internal and external energy. The approach is validated by a series of tests on high resolution satellite images.