Ant-Snake model for Linear Feature Extraction from Satellite Image

Abstract

This paper proposes an optimized mathematical model for linear feature extraction from satellite images. The model is based on a developed ant colony model combined with the snake model (called Ant-Snake model) to identify and extract the linear features like roads from satellite images. The process is started with the developed ant colony model to recognize and identify interest object and then with a snake model extract object. The developed ant model is able to establish a pheromone matrix that represents the object information at each pixel position of the image, according to the movements of a number of ants which are dispatch to move on the image. And the snake model is a parametric curve which is allowed to deform from some arbitrary initial locations from pheromone matrix toward the desired final location by minimizing an energy function. Experimental results are provided to demonstrate the superior performance of the proposed approach.