A new decision criterion for feature selection application to the classification of non destructive testing signatures

European Signal Processing Conference

Volumn 1998-January, Issue , 1998, Pages

  Oukhellou, Latifa ^a   Aknin, Patrice ^a   Stoppiglia, Hervé ^b   Dreyfus, Gérard ^b  

^a Lab des Technologies Nouvelles   (France)

^b PSL RESEARCH UNIVERSITY   (France)

Author keywords

[No Author keywords available]

Indexed keywords

NONDESTRUCTIVE EXAMINATION; PATTERN RECOGNITION; SIGNAL PROCESSING;

CURSE OF DIMENSIONALITY; DECISION CRITERIONS; DIAGNOSIS PROBLEM; NON DESTRUCTIVE TESTING; ORTHOGONAL FORWARD REGRESSIONS; PATTERN RECOGNITION PROBLEMS; RANKING PROCEDURES; REPRESENTATION SPACE;

FEATURE EXTRACTION;

EID: 84890542387     PISSN: 22195491     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (10)

1. Bishop C.M. Neural Networks for Pattern Recognition. Ed Clarendon Press. 1995.
2. Kittler J. Feature Selection and extraction. Handbook of pattern recognition and image processing. Ed. Academic Press. 1986.
3. Chen S. Billings, W.S.A. Luo, Orthogonal least squares methods and their application to non-linear system identification. Int J. Control vol 50 no 5 pp. 1873-1896. 1989.
4. Akaike H. A new look at the statistical model identification. IEEE Trans. on Automatic Control. vol 19 no 6 pp.716-723. 1974.
5. Stoppiglia H. Méthodes Statistiques de Sélection de Modèles Neuronaux. Applications Financières et Bancaires. Thèse de l'Université Pierre et Marie Curie, Paris. 1997.
6. Oukhellou L. Aknin, P.J.P. Perrin, Specific sensor and classifier of defaults of rail head for railway systems. 8th IFAC Symposium on Transportation Systems. Chania. Grèce. 1997.
7. Oukhellou L., Aknin P. Modified Fourier Descriptors : A new parametrization of eddy current signatures applied to the rail defect classification. III Intern. Workshop on Advances in Signal Processing for Non Destructive Evaluation of Materials. Quebec. 1997.
8. Price D., Knerr S., Personnaz L., Dreyfus G. Pairwise neural network classifiers with probabilistic outputs. Neural Information Proc. Systems. 1994.
9. Oukhellou L., Aknin P. Optimization of radial basis function network for classification tasks. Intern. Conference on Neural Networks and theirs Applications. Marseille. 1998.
10. Duprat A.F., Huynh T., Dreyfus G. Towards a Principled Methodology for Neural Network Design and Evaluation in QSAR; Application to the Prediction of Log P. J. Chem. Inf. Comput. Sci., in press (1998).