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...

: This report presents a scene interpretation system in a context of multisensor fusion; it has been applied to the interpretation of remotely sensed images. First we present a typology of the multisensor fusion concepts involved, and we derive the consequences of modeling problems for objects, scene and strategy. The proposed multi-specialist architecture generalizes the ideas of our previous work [GG90a] by taking into account the knowledge about sensors, the multiple viewing notion (shot), and the uncertainty and impresision of models and data modeled with the Possibility Theory. In particular, generic models of objects are represented by concepts independent of sensors (geometry, materials, and spatial context). Three kinds of specialists are present in the architecture: generic specialists (scene and conflict), semantic object specialists, and low level specialists. A blackboard structure with a centralized control is used. The interpreted scene is implemented as a matrix of point...

We present a novel method for the incorporation of shape information into active contour models, and apply it to the extraction of line networks (e.g. road, water) from satellite imagery. The method is based on a new class of contour energies. These energies are quadratic on the space of one-chains in the image, as opposed to classical energies, which are linear. They can be expressed as double integrals on the contour, and thus incorporate non-trivial interactions between different contour points. The new energies describe families of contours that share complex geometric properties, without making reference to any particular shape. Networks fall into such a family, and to model them we make a particular choice of quadratic energy whose minima are reticulated. To optimize the energies, we use a level set approach. The forces derived from the new energies are non-local however, thus necessitating an extension of standard level set methods. Promising experimental results are obtained using real images.

Abstract—This paper addresses the problem of unsupervised extraction of line networks (for example, road or hydrographic networks) from remotely sensed images. We model the target line network by an object process, where the objects correspond to interacting line segments. The prior model, called “Quality Candy, ” is designed to exploit as fully as possible the topological properties of the network under consideration, while the radiometric properties of the network are modeled using a data term based on statistical tests. Two techniques are used to compute this term: one is more accurate, the other more efficient. A calibration technique is used to choose the model parameters. Optimization is done via simulated annealing using a Reversible Jump Markov Chain Monte Carlo (RJMCMC) algorithm. We accelerate convergence of the algorithm by using appropriate proposal kernels. The results obtained on satellite and aerial images are quantitatively evaluated with respect to manual extractions. A comparison with the results obtained using a previous model, called the “Candy ” model, shows the interest of adding quality coefficients with respect to interactions in the prior density. The relevance of using an offline computation of the data potential is shown, in particular, when a proposal kernel based on this computation is added in the RJMCMC algorithm. Index Terms—Stochastic processes, Monte Carlo, simulated annealing, edge and feature detection, remote sensing. æ 1

The detection of roads in satellite images has drawn a lot of attention in the last ten years. Problems of resolution, noise, and image understanding are involved, and one of the best method developed so far is the F* algorithm of Fischler, which is based on dynamic programming. We present herein a mathematical formalization of the F*, an extension to cliques of higher order to deal with the contrast, an extension to neighborhoods of higher order to take into account the curvature, and a physical dynamic model to use the curvature information in a more natural and global way. Keywords edge detection, energy minimization, dynamic programming, curvature, satellite images R'esum'e La reconnaissance de routes sur des images satellite a soulev'e beaucoup d'interet au cours des dix derni`eres ann'ees. Des probl`emes de r'esolution, de bruit, et de compr'ehension de l'image sont impliqu'es et l'une des meilleures m'ethodes d'evelopp'ees jusqu'`a pr'esent est l'algorithme F* de Fischler, qui...

The detection of roads in satellite images has drawn a lot of attention in the last ten years. Problems of resolution, noise, and image understanding are involved, and one of the best method developed so far is the F* algorithm of Fischler, which is based on dynamic programming. We present herein a mathematical formalization of the F*, an extension to cliques of higher order to deal with the contrast, an extension to neighborhoods of higher order to take into account the curvature, and a physical dynamic model to use the curvature information in a more natural and global way. Keywords edge detection, energy minimization, dynamic programming, curvature, satellite images R'esum'e La reconnaissance de routes sur des images satellite a soulev'e beaucoup d'interet au cours des dix derni`eres ann'ees. Des probl`emes de r'esolution, de bruit, et de compr'ehension de l'image sont impliqu'es et l'une des meilleures m'ethodes d'evelopp'ees jusqu'`a pr'esent est l'algorithme F* de Fischler, qui...

In this paper we propose a new method to extract roads in remote sensed images. Our approach is based on stochastic geometry theory and reversible jump Monte Carlo Markov Chains dynamic. We consider that roads consist of a thin network in the image. We make the hypothesis that such a network can be approximated by a network composed of connected line segments. We build a marked point process, which is able to simulate and detect thin networks. The segments have to be connected, in order to form a line-network. Aligned segments are favored whereas superposition is penalized. Those constraints are taken in account by the prior model (Candy model), which is an area-interaction point process.The location of the network and the specities of a road network in the image are given by the likelihood term. This term is based on statistical hypothesis tests. The proposed probabilistic model yelds a MAP estimator of the road network. In order to avoid local minima, a simulated annealing algori...

Satellite image classification has been a major research field for many years with its varied applications in the field of Geography, Geology, Archaeology, Environmental Sciences and Military purposes. Many different techniques have been proposed to classify satellite images with color, shape and texture features. Complex indices like Vegetation index (NDVI), Brightness index (BI) or Urban index (ISU) are used for multi-spectral or hyper-spectral satellite images. In this paper we will show the efficiency of structural features describing man-made objects in mid-resolution satellite images to describe image content. We will then show the state-of-the-art to classify large satellite images with structural features computed from road networks and urban regions extracted on small image patches cut in the large image. Fisher Linear Discriminant (FLD) analysis is used for feature selection and a one-vsrest probabilistic Gaussian kernel Support Vector Machines (SVM) classification method is used to classify the images. The classification probabilities associated with each subimage of the large image provide an estimate of the geographical class coverage. 1.