This Article 
 Bibliographic References 
 Add to: 
An Active Testing Model for Tracking Roads in Satellite Images
January 1996 (vol. 18 no. 1)
pp. 1-14

Abstract—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 minimization problem is solved (exactly), resulting in a new image location to inspect, and so forth. We report experiments using panchromatic SPOT satellite imagery with a ground resolution of ten meters: Given a starting point and starting direction, we are able to rapidly track highways in southern France over distances on the order of one hundred kilometers without manual intervention.

[1] E. Arkin, H. Meijer, J. Mitchell, D. Rappaport, and S. Skiena, "Decision trees for geometric models," Proc. Ninth ACM Symp. Computational Geometry, 1993.
[2] M. Barzohar and D.B. Cooper, "Completely automatic reliable finding of main roads in aerial imagery by using Bayesian methods," Technical Report, Brown-LEMS-118, Mar. 1993.
[3] H.V. Blanc, "Application du groupement perceptuelàla reconnaissance de routes sur une image satellite SPOT," Technical Report, Univ. de Montpellier II Sciences et Techniques du Languedoc, 1993.
[4] L. Breiman, J. Friedman, R. Olshen, and C. Stone, Classification and Regression Trees.Belmont, Calif.: Wadsworth, 1984.
[5] J. Boggess, "Identification of Roads in Satellite Imagery Using Artificial Neural Networks: A Contextual Approach," technical report, Mississippi State Univ., Aug. 1993.
[6] H. Chernoff, Sequential Analysis and Optimal Design.Philadelphia: SIAM, 1972.
[7] P. Chou,“Optimal partitioning for classification and regression trees,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 13, no. 4, pp. 340-354, Apr. 1991.
[8] M. Daoud, C. Roux, and A. Hillion, "Une application de la théorie des graphesàl'extraction automatique des réseaux de communication dans les images SPOT," GRETSI, pp. 699-701, June 1989.
[9] R.O. Duda and P.E. Hart, Pattern Classification and Scene Analysis,New York: John Wiley&Sons, 1973.
[10] I. Estival and H. Le Men, "Detection of linear networks on satellite images," Proc. Conf. Pattern Recognition, pp. 856-858, Oct. 1986.
[11] M. Fischler, J. Tenenbaum, and H. Wolf, "Detection of roads and linear structures in low-resolution aerial imagery using a multisource knowledge integration technique," Computer Graphics and Image Processing, vol. 15, pp. 201-223, 1981.
[12] M.R. Garey, "Optimal binary identification procedures," SIAM J. Applied Mathematics, vol. 23, pp. 173-186, 1972.
[13] M.R. Garey and R.L. Graham, "Performance bounds on the splitting algorithm for binary testing," Acta Informatica, vol. 3, pp. 347-355, 1974.
[14] P. Garnesson, "MESSIE: un système d'analyse de scènes," Technical Report, Univ. de Nice-Sophia Antipolis, 1991.
[15] D. Geman and B. Jedynak, "Detection of roads in satellite images," Proc. IGARSS, Espoo, 1991.
[16] D. Geman and B. Jedynak, "Shape recognition and twenty questions," Technical Report No. 2155, INRIA-Rocquencourt, Nov. 1993.
[17] J.C. Gittins, Multi-armed Bandit Allocation Indices.New York: John Wiley&Sons, 1989.
[18] C. Goad, "Special purpose automatic programming for three- dimensional model-based vision," Proc. ARPA Image Understanding Workshop, pp. 94-104, 1983.
[19] C. Graffigne and I. Herlin, "Modélisation de réseaux pour l'imagerie satellite SPOT," Technical Report, INRIA, 1989.
[20] Y.X. Gu, Q.R. Wang, and C.Y. Suen, "Application of multilayer decision tree in computer recognition of Chinese characters," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 5, pp. 83-89, 1983.
[21] C. Hansen and T. Henderson, "Towards the automatic generation of recognition strategies," Second Int'l Conf. Computer Vision, IEEE, pp. 275-279, 1988.
[22] R.M. Haralick and L.G. Shapiro, Computer and Robot Vision. New York: Addison-Wesley, 1993.
[23] C. Hartmann, P. Varshney, K. Mehrotra, and C. Gerberich, "Application of Information Theory to the Construction of Decision Trees," IEEE Trans. Information Theory, July 1982.
[24] D.A. Huffman, "A method for the construction of minimum redundancy codes," Proc. I.R.E., vol. 40, pp. 1,098-1,101, 1952.
[25] L. Hyafil and R. Rivest, "Constructing optimal binary decision trees is NP-complete," Information Processing Letters, vol. 5, pp. 15-17, 1976.
[26] B. Jedynak, "Modèles stochastiques et méthodes déterministes pour extraire les routes des images de la Terre vues du ciel" Thèse de Doctorat, Modèlisation Stochastique, Univ. Paris Sud, 1995.
[27] I. Keller, "Recherche d'un meilleur modèle a priori pour une méthode d'extraction des routes dans une image satellite," Technical Report, Univ. de Paris-Sud (Orsay), 1994.
[28] P.R. Kumar, "A survey of some results in stochastic adaptive control," SIAM J. Control and Optimization, vol. 23, pp. 329-380, 1985.
[29] M.W. Kurzynski, "The optimal strategy of a tree classifier," Pattern Recognition, vol. 16, pp. 81-87, 1983.
[30] E. Lawler, Combinatorial Optimization: Networks and Matroids. Saunders College Publishing, 1976.
[31] Y.K. Lin and K.S. Fu, "Automatic classification of cervical cells using a binary tree classifier," Pattern Recognition, vol. 16, pp. 69-80, 1983.
[32] J.C. Newell, D.G. Gisser, and D. Isaacson, “An Electric Current Tomograph,” IEEE Trans. Biomedical Eng., Vol. 35, 1988, pp. 828-833.
[33] N. Merle and J. Zerubia, "A curvature-dependent energy function for detecting lines in satellite images," Eighth SCIA Conf., May, 1993.
[34] M. Miyakawa, "Criteria for selecting a variable in the construction of efficient decision trees," IEEE Trans. Computers, vol. 38, pp. 130-141, 1989.
[35] R. Neviata, "Locating structures in aerial images," IEEE Trans. Pattern Analysis and Machine Intelligence, pp. 476-484, Sept. 1982.
[36] J.R. Quinlan,"Induction of decision trees," Machine Learning, vol. 1, pp. 81-106, 1986.
[37] R.D. Rimey and C.M. Brown, "Control of Selective Perception Using Bayes Nets and Decision Theory," Int'l J. Computer Vision, vol. 12, pp. 173-207, 1994.
[38] M. Roux, "Recalage d'images multi-sources. Application au recalage d'une image SPOT et d'une carte," Thèse de Doctorat, Ecole Nationale Supèrieure des Télécommunications, Sept. 1992.
[39] M. Sandelius, "On an optimal search procedure," Am. Math. Monthly, vol. 68, pp. 133-134, 1961.
[40] M.A. Serendero, "Extraction d'informations symboliques en imagerie SPOT: réseaux de communication et aglomations," Thèse de Doctorat, Univ. de Nice, 1989.
[41] L. Spirkovska, "Three-dimensional object recognition using similar triangles and decision trees," Pattern Recognition, vol. 26, pp. 727-732, 1993.
[42] M. Swain, "Object recognition from a large database using a decision tree," Proc. DARPA Image Understanding Workshop, Morgan Kaufman, 1988.
[43] M.J. Swain and M.A. Stricker, “Promising Directions in Active Vision,” Int'l J. Computer Vision, vol. 11, no. 2, pp. 109-126, 1993.
[44] G.J. Vanderbrug, "Line detection in satellite imagery," IEEE Trans. Geoscience Electronics, vol. 14, pp. 37-44, Jan. 1976.
[45] F. Wang and R. Newkirk, "A knowledge-based system for highway network extraction," IEEE Trans. Geoscience and Remote Sensing, vol. 26, Sept. 1988.
[46] Q.R. Wang and C.Y. Suen, "Analysis and design of a decision tree based on entropy reduction and its application to large character set recognition," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 6, pp. 406-417, 1984.
[47] S. Watanabe, "Pattern recognition as a quest for minimum entropy," Pattern Recognition, vol. 13, pp. 381-387, 1981.
[48] S. Zimmerman, "An optimal search procedure," Am. Math. Monthly, vol. 66, pp. 690-693, 1959.

Index Terms:
Decision tree, model-based tracking, active testing, roads, SPOT images.
Donald Geman, Bruno Jedynak, "An Active Testing Model for Tracking Roads in Satellite Images," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 18, no. 1, pp. 1-14, Jan. 1996, doi:10.1109/34.476006
Usage of this product signifies your acceptance of the Terms of Use.