Abstract
A technique is proposed for choosing the thresholds for a number of object detection tasks, based on a prototype selection technique. The chosen prototype subset has to be correctly classified. The positive and negative objects are introduced in order to provide the optimization via empirical risk minimization. A Boolean function and its derivatives are obtained for each object. A special technique, based on the fastest gradient descent, is proposed for the sum of Boolean functions maximization. The method is applied to the detection task of house edges, using its images in aerial photos. It is shown that proposed method can be expanded to solving of a wide range of tasks, connected to the function optimization, while the function is given in vertices of a 2n single hyper - cube.
Chapter PDF
Similar content being viewed by others
References
Antamoshkin, A. A., Saraev, V. M.: On Definition of Informative Subsystem of Signs in the Pattern Recognition Problem. Computers and Artificial Intell. 4 (1985) 245–252
Beveridge J. R., Riserman E. M.: How Easy Matching 2D Line Models Using Local Search? IEEE Trans, on Pattern Anal, and Machine Intell. 6(1997) 564–579
Decaestecker C: Finding Prototypes for Nearest Neighbor Classification by Means of Gradient Descent and Deterministic Annealing. Pattern Recognition 30(1997) 281–288
Demigny D., Kamle T.: A Discrete Expression of Canny’s Criteria for Step Edge Detector Performances Evaluation. IEEE Trans, on Pattern Anal, and Machine Intell. 19(1977) 1199–1211
Dennis J. E. Jr., Schnabel R. B.: Numerical Methods for Unconstrained Optimization and Nonlinear Equations. Prentice-Hall Inc., New Jersey (1983)
Devijer P. A., Kittler J.: Pattern Recognition: A Statistical Approach. Prentice-Hall Inc., Englewood Cliffs, N. J. (1982)
Foroutan I., Sklansky J.: Feature Selection for Automatic Classification of Non-Gaussian Data. IEEE Trans, on System Man and Cybernetics 17(1987) 187–198
Gates G. W.: The Reduced Nearest Neighbor Rule. Trans, on Inf. Theory 18(1972) 431–433
Hart P. E.: The Condensed Nearest Neighbor Rule. Trans on Inf. Theory 14(1968) 515–516
Jain A., Zongker D.: Feature Selection: Evaluation, Application and Small Sample Performance. IEEE Trans. on Pattern Anal. and Mach. Intell. 19(1997) 153–158
Kim J., Yu J. R., Kim S. H.: Learning of Prototypes and Decision Boundaries for a Verification Problem. Pattern Recognition Lett. 17(1996) 691–697
Kittler J.: Feature Set Search Algorithms. In Chen C. H. (ed.): Pattern Recognition and Signal Processing. Sijthoff and Noordhoff, Alphen van der Rijn, The Nethrlands (1978) 41–60
Kuncheva L. I.: Editing For the k-Nearest Neighbors Rule by a Genetic Algorithm. Pattern Recognition Lett. 16(1995) 809–814
Lipowezky U.: Tree-Plantation Decipherment of Panchromatic Aerial Photo Images Using Supervised Template Matching. In Proc. of 9-th. Mediterranean Electromechanical Conf. Melecon’98, vol. 1 Tel-Aviv (1998) 48–52
Lipowezky U.: Selection of the Optimal Prototype Subset for 1-NN Classification. Pattern Recognition Lett. 19(1998) 907–918
Sanchez J. S., Fla F., Ferry F. J.: Prototype Selection for the Nearest Neighbor Rule Through Proximity Graph. Pattern Recognition Lett. 18(1997) 507–513
Strela V., Heller P. N., Strang G., Topiwala P., Heil C: The Application of Multi-wavelet Filterbanks for Image Processing. IEEE Trans. on Image Processing 8(1999) 548–563
Trier Ø. D., Jain A. K.: Goal-Directed Evaluation of Binarization Methods. IEEE Trans, on Pattern Anal. and Mach. Intel. 17(1995) 1191–1201
Vapnik V. N.: Estimation of Dependencies Based on Empirical Data. Springer-Verlag, Berlin Heidelberg (1982)
Zadeh L. A.: Fuzzy Sets as a Basis for a Theory of Possibility. Fuzzy Sets Syst. 1(1978) 3–28
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2000 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Lipowezky, U., Shenkar, V. (2000). Optimal Threshold Estimation Using Prototype Selection. In: Ferri, F.J., Iñesta, J.M., Amin, A., Pudil, P. (eds) Advances in Pattern Recognition. SSPR /SPR 2000. Lecture Notes in Computer Science, vol 1876. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-44522-6_76
Download citation
DOI: https://doi.org/10.1007/3-540-44522-6_76
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-67946-2
Online ISBN: 978-3-540-44522-7
eBook Packages: Springer Book Archive