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...

Humans can fairly easily identify roads in remote sensing images, but this has turned out to be a difficult task for computers. Most previous work in this area has utilized statistical and rule-based techniques, which depended primarily upon spectral information. However, it appears that spectral information alone is insufficient to identify roads in Landsat Thematic Mapper satellite imagery, since soils have the same spectral signature in the data as roads, and that contextual information is required. In this application, artificial neural networks are found to be superior to several previous techniques due in part to their ability to incorporate both spectral and contextual information. However, a number of factors cause problems for the network, and further work must be done to include additional information; it is suggested that a hybrid system might alleviate most of these difficulties. Key words: Artificial neural networks, Satellite imagery classification, Artificial intelligenc...

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

This paper addresses the issue of extracting main and secondary road networks in dense urban areas from very high resolution (VHR, ~0.61m) satellite images. The difficulty with secondary roads lies in the low discriminative power of the grey-level distributions of road regions and the background, and the greater effect of occlusions and other noise on narrower roads. To tackle this problem, we use a previously developed higher-order active contour (HOAC) phase field model and augment it with an additional non-linear non-local term. The additional term allows separate control of road width and road curvature; thus more precise prior knowledge can be incorporated, and better road prolongation can be achieved for the same width. Promising results on QuickBird panchromatic images at reduced resolutions and comparisons with other models demonstrate the role and the efficiency of our new model. 1.

www.elsevier.com/locate/isprsjprs Extraction of tidal channel networks from airborne scanning laser altimetry

Abstract not found