SINGLE VALUES SELECTIONS FOR FILTERING IMAGE
DOI:
https://doi.org/10.26512/2236-56562009e39845Keywords:
Single Values Decomposition, Self-Organizing Maps, Filtering, Image ProcessingAbstract
This work explores the spatial relation of Single Values over image pixels in order to propose a filtering process. Different denoise degrees are applied to each pixel considering their effects on an unsupervised clustering process – Self Organizing Map. This results on specific filtering process to different spectral characteristics of the images. Two experiments are presented; the first one with synthetic data and the second with LandSat-7 data. The first one considers a scene with high frequencies and consequently cuts only the null space of the single values of each pixel. The experiment with LandSat-7 data shows a case with homogeneous scenes. In this case, the filtering process implements hard cuts considering a limited group of classes. The technique presented here brings an effective way to reconstruct better approximations of the original data and, at the same time, excludes unnecessary ranges of pixel variations.
References
HAYKIN, S. (1998). Neural Networks – A comprehensive foundation. Prentice Hall; 2 edition, 842 p.
KOHONEN. T. (1990). Speech recognition based on topology-preserving neural maps. In: ALEKSANDER, I. (Org.). Neural Computing Architectures, Massachusetts: MIT Press, Cambridge, p. 26-40.
MALSBURG, C. VON DER (1990). Network self-organization. In: ZORNETZER, S.F.; DAVIS, J.L. & LAU, C (Org.). An introduction to neural and electronic networks. San Diego, CA: Academic Press, p. 421-432.
MEDEIROS F. N. S, MASCARENHAS, N. D. A. & COSTA, L. F. (1998). Adaptive Speckle MAP Filtering for SAR Images Using Statistical Clustering. In: International Symposium on Computer Graphics, Image Processing, and Vision. Rio de Janeiro, Brazil. Proceedings SIBGRAPI’98, 303 – 310.
NOBLE, B. & DANIEL, J. W. (1986). Álgebra linear aplicada. 2ª ed. Rio de Janeiro: Prentice-Hall do Brasil, 262-280.
RITTER, H. & SCHULTEN, K. (1988). Convergence properties of Kohonen’s topology conserving maps: Fluctuations, stability and dimension selection. Biological Cybernetics, 60: 59-71.
SANT’ANNA, S. J. S. & MASCARENHAS, N. D. A. (1994). Avaliação comparativa da perda de resolução espacial de filtros redutores de ruído speckle. In: Simpósio Brasileiro de Computação Gráfica e Processamento de Imagens, 7., Curitiba, PR. Anais. Curitiba: SBC/UFPR, 141-148.
WEIGANG, L & SILVA, N. C, (1999). Implementation of parallel self-organizing map to the classification of the image. Proc. SPIE- Int. Soc. Opt. Eng. 3722 , p. 284-292.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
