Artículos

 

Is the Coordinated Clusters Representation an analog of the Local Binary Pattern?

 

¿La Representación de Imágenes por Cúmulos Coordinados es análogo al de Patrones Binarios Locales?

 

E.V. Kurmyshev

 

Centro de Investigaciones en Óptica, A.C., Loma del Bosque 115, Lomas del Campestre, 37150 León, Guanajuato, México. E–mail: kev@cio.mx

 

Article received on September 22, 2008
Accepted on March 13, 2009

 

Abstract

Both the Local Binary Pattern (LBP) and the Coordinated Clusters Representation (CCR) are two methods used successfully in the classification and segmentation of images. They look very similar at first sight. In this work we analyze the principles of the two methods and show that the methods are not reducible to each other. Topologically they are as different as a sphere and a torus. In extracting of image features, the LBP uses a specific technique of binarization of images with the local threshold, defined by the central pixel of a local binary pattern of an image. Then, the central pixel is excluded of each local binary pattern. As a consequence, the mathematical basis of the LBP method is more limited than that of the CCR. In particular, the scanning window of the LBP has always an odd dimensions, while the CCR has no this restriction. The CCR uses a binarization as a preprocessing of images, so that a global or a local threshold can be used for that purpose. We show that a classification based on the CCR of images is potentially more versatile, even though the high performance of both methods was demonstrated in various applications.

Keywords: Texture Image Analysis, Classification, Segmentation, Coordinated Clusters Representation, Local Binary Patterns.

 

Resumen

La Representación de Imágenes por Cúmulos Coordinados (RICC) y el Local Binary Pattern (LBP) son métodos eficazmente usados para la clasificación y segmentación de imágenes. A primera vista éstos parecen muy similares. Con un análisis de los principios de dos métodos demostramos que no son reducibles uno a otro; en términos de topología matemática son tan diferentes como esfera y dona. En la etapa de extracción de características de una imagen, el LBP usa una técnica específica de binarización de imágenes con umbral local, que se define por el píxel central de un patrón local de la imagen. Después, el píxel central se excluye de cada patrón local. Por tanto, el sustento matemático del método de LBP es más limitado que el de la RICC. En particular, la ventana de barrido en LBP siempre tiene dimensiones impares, la de la RICC no tiene esta restricción. La RICC requiere la binarización como una etapa de preprocesado de imagen y, por tanto, puede usarse un umbral global o local adaptable. La clasificación basada en la RICC es más versátil, aunque las eficiencias terminales de clasificación por los dos métodos pueden ser muy cercanas en muchas aplicaciones.

Palabras clave: Análisis de Imágenes de Textura, Clasificación, Segmentación, Representación de Imágenes por Cúmulos Coordinados, Patrones Binarios Locales.

 

DESCARGAR ARTÍCULO EN FORMATO PDF

 

References

1. Azencott, R., & Wang, J.P., (1997). Texture classification using windowed Fourier filters. IEEE Transactions on Pattern Analysis and Machine Intelligence, 19(2), 148–153.         [ Links ]

2. Berry, J.R & Goutsias, J., (1989). Comparison of Matrix Texture Features Using a Maximum Likelihood Texture Classifier. SPIE, Visual Communications and Image Processing IV, Pennsylvania, USA, vol. 1199, 305–316.         [ Links ]

3. Chellappa, R., & Chatterjee, S., (1987). Classification of textures using Gaussian Markov random fields. IEEE Transactions on Acoustics, Speech, and Signal Processing, 33 (4), 959–963.         [ Links ]

4. Chen, C.H., Pau, L.F., & Wang, P. S. P. (1996). Handbook of Pattern Recognition & Computer Vision. Singapore: World Scientific.         [ Links ]

5. Chetverikov, D. (1999). Texture analysis using feature–based pair wise interaction maps. Pattern Recognition, 32(3), 487–502.         [ Links ]

6. Duda, R.O., Hart, P.E., & Stork, D.G. (2001). Pattern Classification, (2nd ed.) New York: John Wiley & Sons Inc.         [ Links ]

7. Elfadel, I.M., & Picard, R.W. (1994). Gibbs random fields, co–occurrences and texture modeling. IEEE Transactions on Pattern Analysis and Machine Intelligence, 16(1), 24–37.         [ Links ]

8. Fukunaga, K. (1990). Introduction to Statistical Pattern Recognition, (2nd ed.) New York: Academic Press.         [ Links ]

9. Gonzalez–Garcia, A.C., Sossa–Azuela, J.H., Felipe–Riveron, E.M. & Pogrebniak, O. (2007). Image Retrieval Based on Wavelet Transform and Neural Network Classification, Computación y Sistemas, 11(2), 143–156.         [ Links ]

10. Goon, A., & Rolland, J.P. (1999). Texture classification based on comparison of second–order statistics I: 2P–PDF estimation and distance measure. Journal of Optical Society of America A, 16 (7), 1566–1574.         [ Links ]

11. Haralick, R.M. (1979). Statistical and structural approaches to texture. Proccedings of the IEEE, 67(5), 786–804.         [ Links ]

12. Kurmyshev, E.V., Gusakov, V.E., & Sissakian, I.N. (1987). Coordinated clusters representation for non–crystalline solids. Moscow, Russia: General Physics Institute.         [ Links ]

13. Kurmyshev, E.V., & Cervantes, M. (1996). A quasi–statistical approach to digital image representation. Revista Mexicana de Física, 42 (1), 104–116.         [ Links ]

14. Kurmyshev, E.V., & Soto, R. (1996). Digital pattern recognition in the coordinated cluster representation. Nordic Signal Processing Simposium, Espoo, Finland, 463–466.         [ Links ]

15. Kurmyshev, E.V., & Sánchez–Yáñez, R.E. (2001). Texture classification based on image representation by coordinated clusters. Image and Vision Computing'2001 New Zealand, Dunedin, New Zealand, 213–217.         [ Links ]

16. Kurmyshev, E.V., Cuevas, F.J., & Sánchez Yáñez, R.E. (2003). Noisy binary texture recognition using the coordinated cluster transform. Computación y Sistemas, 6 (3), 196–203.         [ Links ]

17. Kurmyshev, E.V., & Sánchez–Yáñez, R.E. (2005). Comparative experiment with color texture classifiers using the CCR feature space. Pattern Recognition Letters, 26 (9), 1346–1353.         [ Links ]

18. Manjunath, B.S., & Ma, W.Y. (1996). Texture features for browsing and retrieval of image data. IEEE Transactions on Pattern Analysis and Machine Intelligence, 18 (8), 837–842.         [ Links ]

19. Maenpaa, T. (2003). The local binary pattern approach to texture analysis – Extensions and applications. PhD Thesis, University of Oulu, Oulu, Finland.         [ Links ]

20. Maenpaa, T., & Pietikainen, M. (2004). Classification with color and texture: Jointly or separately? Pattern Recognition, 37(8), 1629–1640.         [ Links ]

21. Ohanian, P.P., & Dubes, R.C. (1992). Performance evaluation for four classes of textural features. Pattern Recognition, 25(8), 819–833.         [ Links ]

22. Ojala, T., Pietikainen, M., & Harwood, D. (1996). A comparative study of texture measures with classification based on feature distributions. Pattern Recognition, 29(1), 51–59.         [ Links ]

23. Ojala, T., Pietikainen, M., & Nisula, J. (1996). Determining composition of grain mixtures by texture classification based on feature distributions. International Journal of Pattern Recognition and Artificial Intelligence, 10(1), 73–82.         [ Links ]

24. Ojala, T., Valkealahti, K., & Pietikainen, M. (2000). Texture discrimination with multidimensional distributions of signed gray level differences. Pattern Recognition, 34(1), 21–33.         [ Links ]

25. Ojala, T., Pietikainen, M., & Maenpaa, T. (2002). Multiresolution Gray–Scale and Rotation Invariant Texture Classification with Local Binary Patterns. IEEE Transactions on Pattern Analysis and Machine Intelligence, 24(7), 971–987.         [ Links ]

26. Pietikainen, M., Ojala, T., & Xu, Z. (2000). Rotation–invariant texture classification using feature distributions, Pattern Recognition, 33(1), 43–52.         [ Links ]

27. Sánchez Yáñez, R.E., Kurmyshev, E.V., & Cuevas, F.J. (2003). A framework for texture classification using the coordinated clusters representation. Pattern Recognition Letters. 24 (13), 21–31.         [ Links ]

28. Sánchez Yáñez, R.E., Kurmyshev, E.V., & Fernández, A. (2003). One–class texture classifier in the CCR feature space. Pattern Recognition Letters 24 (9–10), 1503–1511.         [ Links ]

29. Kurmyshev, E.V. (2007). Classification of texture images using coordinated clusters representation, Recent Advances in Optical Metrology, 155–226. Kerala, India: Research Signpost. Ordenar alfabéticamente esta referencia.         [ Links ]

30. Soh, L.K., & Tsatsoulis, C. (1999). Texture analysis of SAR sea ice imagery using gray level co–occurrence matrices. IEEE Transactions on Geosciences and Remote Sensing, 37(2), 780–795.         [ Links ]

31. Tuceryan, M., & Jain, A. K. (1993). Texture analysis. In Chen, C.H., Pau, L.F., Wang, P.S.P. (Eds.), Handbook of Pattern Recognition and Computer Vision. (235–276), Singapore: World Scientific Publishing Company.         [ Links ]

32. Turner, M.R. (1986). Texture discrimination by Gabor functions, Biological Cybernetics, 55 (2–3), 71–82.         [ Links ]

33. Valkealahti, K., & Oja, E. (1998). Reduced multidimensional co–occurrence histograms in texture analysis. IEEE Transactions on Pattern Analysis and Machine Intelligence, 20(1), 90–94.         [ Links ]

34. Wang, L., & He, D.C. (1990). Texture classification using texture spectrum, Pattern Recognition, 23(8), 905–910.         [ Links ]

35. Young, T. Y., & Fu, K.–S. (Eds.), (1986). Handbook of Pattern Recognition and Image Processing. Orlando: Academic Press.         [ Links ]