DATA DIMENSIONALITY REDUCTION FOR NEURAL BASED CLASSIFICATION OF OPTICAL SURFACES DEFECTS
DOI:
https://doi.org/10.47839/ijc.8.1.654Keywords:
Computer Aided Diagnosis Systems (CADS), Artificial Intelligent systems, Industrial applications, Artificial Neural Network, Dimensionality Reduction, Curvilinear Component Analysis (CCA), Curvilinear Distance Analysis (CDA), Self Organizing Maps (SOM).Abstract
A major step for high-quality optical surfaces faults diagnosis concerns scratches and digs defects characterization in products. This challenging operation is very important since it is directly linked with the produced optical component’s quality. A classification phase is mandatory to complete optical devices diagnosis since a number of correctable defects are usually present beside the potential “abiding” ones. Unfortunately relevant data extracted from raw image during defects detection phase are high dimensional. This can have harmful effect on the behaviors of artificial neural networks which are suitable to perform such a challenging classification. Reducing data dimension to a smaller value can decrease the problems related to high dimensionality. In this paper we compare different techniques which permit dimensionality reduction and evaluate their impact on classification tasks performances.References
M. Voiry, F. Houbre, V. Amarger, and K. Madani. Toward Surface Imperfections Diagnosis Using Optical Microscopy Imaging in Industrial Environment. Proceedings of the Workshop IAR & ACD 2005, Mulhouse, France 16-18 November 2005, pp. 139-144.
M. Voiry, V. Amarger, K. Madani, and F. Houbre. Combining Image Processing and Self Organizing Artificial Neural Network Based Approaches for Industrial Process Faults Clustering. Proceedings of 13th International Multi-Conference on Advanced Computer Systems (ACS 2006), Miedzyzdroje, Poland 18-20 October 2006, pp. 129-138.
M. Voiry, K. Madani, V. Amarger, and F. Houbre. Toward Automatic Defects Clustering in Industrial Production Process Combining Optical Detection and Unsupervised Artificial Neural Network Techniques. Procedings of the 2nd International Workshop on Artificial Neural Networks and Intelligent Information Processing (ANNIIP 2006) ,Setubal, Portugal August 2006, pp. 25-34.
G. P. Zhang. Neural Networks for Classification: A Survey. IEEE Trans. on Systems, Man, and Cybernetics - Part C: Applications and Reviews 30 (4) (2000). p. 451-462.
M. Egmont-Petersen, D. de Ridder, and H. Handels. Image Processing with Neural Networks - A Review. Pattern Recognition 35 (2002). p. 2279-2301.
P. J. Grother. Karhunen Loeve Feature Extraction for Neural Handwritten Character Recognition. Proceedings of SPIE, vol. 1709, no. Applications of Artificial Neural Networks III, pp. 155-166, 1992.
K. Boehm, W. Broll, and M. Sokolewicz. Dynamic Gesture Recognition Using Neural Networks; A Fundament for Advanced Interaction Construction. Proceedings of SPIE, vol. 2177, no. Stereoscopic Displays and Virtual Reality Systems, pp. 336-346, 1994.
J. Lampinen and E. Oja. Distortion Tolerant Pattern Recognition Based on Self-Organizing Feature Extraction. IEEE Trans. On Neural Networks 6 (3) (1995). p. 539-547.
S. Buchala, N. Davey, T. M. Gale, and R. J. Frank. Analysis of Linear and Nonlinear Dimensionality Reduction Methods for Gender Classifcation of Face Images. International Journal of Systems Science 14 (36) (2005). p.931-942.
M. Verleysen. Learning high-dimensional data. NATO Advanced Research Workshop on Limitations and Future Trends in Neural Computing (LFTNC'2001), Siena, Italy 22-24 October 2001, pp.22-24.
M. Lennon, G. Mercier, M. C. Mouchot, and L. Hubert-Moy. Curvilinear Component Analysis for Nonlinear Dimensionality Reduction of Hyperspectral Images. Proceedings of SPIE, vol. 4541,Image and Signal Processing for Remote Sensing VII, pp. 157-168, 2001.
P. Demartines. Analyse de Donnees par Reseaux de Neurones Auto-Organises. PhD Thesis Institut National Polytechnique de Grenoble, 1994.
T. Kohonen. Self Organizing Maps. 3rd edition ed. Berlin: Springer, 2001.
T. Kohonen, E. Oja, O. Simula, A. Visa, and J. Kangas. Engineering Applications of the Self-Organizing Maps. Proceedings of the IEEE, vol. 84, no. 10, pp. 1358-1384, Oct.1996.
J. Heikkonen and J. Lampinen. Building Industrial Applications with Neural Networks. Proceedings of European Symposium on Intelligent Techniques (ESIT'99), Chania, Greece June 1999.
P. Demartines and J. Herault. Vector Quantization and Projection Neural Network. Lecture Notes in Computer Science 686, International Workshop on Artificial Neural Networks (IWANN’93), p. 328-333.
P. Demartines and J. Herault.CCA : “Curvilinear Component Analysis”. Proceedings of 15th workshop GRETSI (GRETSI’95), Juan-les-pins, France, 15 September 1995.
J. A. Lee, A. Lendasse, N. Donckers, and M. Verleysen. A Robust Nonlinear Projection Method. Proceedings of European Symposium on Artificial Neural Networks (ESANN'2000).
S. Derrode. Representation de Formes Planes a Niveaux de Gris par Differentes Approximations de Fourier-Mellin Analytique en vue d'Indexation de Bases d'Images. Phd Thesis Universite de Rennes I, 1999.
F. Ghorbel. A Complete Invariant Description for Gray Level Images by the Harmonic Analysis Approach. Pattern Recognition 15 (1994). p. 1043-1051.
P. Demartines and F. Blayo. Kohonen Self-Organizing Maps: Is the Normalization Necessary? Complex Systems 6 (2) (1992). p. 105-123.
P. Grassberger and I. Procaccia. Measuring the strangeness of strange attractors. Physica 9 (1983). p. 189-208.
F. Camastra and A. Vinciarelli. Intrinsic Dimension Estimation of Data: An Approach Based on Grassberger-Procaccia's Algorithm. Neural Processing Letters 14 (1) (2001). p. 27-34.
Downloads
Published
How to Cite
Issue
Section
License
International Journal of Computing is an open access journal. Authors who publish with this journal agree to the following terms:• Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
• Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
• Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
