We present a new approach for tracking roads from satellite images, and thereby illustrate a general computational strategy ("active testing") for tracking 1D structures and other recognition tasks in computer vision. Our approach is related to recent work in active vision on "where to look next" and motivated by the "divide-and-conquer" strategy of parlor games such as "Twenty Questions." We choose "tests" (matched filters for short road segments) one at a time in order to remove as much uncertainty as possible about the "true hypothesis" (road position) given the results of the previous tests. The tests are chosen on-line based on a statistical model for the joint distribution of tests and hypotheses. The problem of minimizing uncertainty (measured by entropy) is formulated in simple and explicit analytical terms. To execute this entropy testing rule we then alternate between data collection and optimization: at each iteration new image data are examined and a new entropy minimizat...

Dpartement de mathmatiques

In this paper, we present a new method for automatic road tracking and extraction from satellite images. Our method is a kind of global and parallel method. It includes the following three parts: I) an edge detector is applied and the satellite image is transfered into binary edge map (EM). 2) we propose a special kind of energy function with marginal effect to represent each edge chain in the EM. Through this energy function, we are able to track and extract the candidates of roads. 3) By applying the generic knowledge of roads, we can easily discard the false candidates, and then produce the road boundaries from the remaining chains. The experiments show that our method can extract the roads from the satellite images robustly and quickly. It is also shown that our method can be extended to extract the linear features in other kinds of images. 1