General Algorithm for the Semantic Decomposition of Geo–Images

Un algoritmo general para la descomposición semántica de Geo–Imágenes

José Giovanni Guzmán Lugo
Graduated on december 4, 2007
Centro de Investigación en Computación, ]]> IPN México D.F., México.
jguzmanl@cic.ipn.mx

Advisor: Serguei Levachkine
Centro de Investigación en Computación,
IPN México D.F., México
sergei@cic.ipn.mx

Abstract

The thesis presents an object oriented methodology for the semantic extraction of a geo–image which is defined by a set of natural language labels. The approach is composed of two main stages: analysis and synthesis. The analysis stage detects the main geographic components of a geo–image by means of the color quantification, geometry and topology of the geospatial objects. The result of this stage is a set of geo–images with intensities that are approximately uniform. The synthesis stage extracts the main geographic objects that have been identified and a labeling process in two levels (general and specialized), which is equivalent to consider both local and global information of a geo–image. The aim of the general labeling process is to associate a label of the adequate thematic to each region, taking into account the RGB characteristics of the image. In order to specialize each geographic object, we have proposed a specialization algorithm that considers geometric and topologic relations among them, represented in geographic application domain ontology. The obtained set of labels describes the geo–image semantics.

Keywords: Image Processing and Computer Vision, Scene Analysis, Object Recognition.

Resumen

Esta tesis presenta una metodología orientada a objetos para la extracción de la semántica de una geo–imagen definida por un conjunto de etiquetas en lenguaje natural. La metodología está compuesta de dos grandes etapas: análisis y síntesis. La etapa de análisis detecta los principales elementos geográficos de una geo–imagen mediante la cuantificación de características como color, geometría y topología de los objetos geográficos. El resultado de esta etapa es un conjunto de geo–imágenes con intensidades de color aproximadamente uniforme. La etapa de síntesis extrae los objetos geográficos que fueron identificados y realiza un proceso de etiquetado en dos niveles (general y especializado), el cual es equivalente a considerar tanto la información global como local de una geo–imagen. El propósito del etiquetado general es asociar a cada región una etiqueta de una temática adecuada, tomando en consideración la información RGB de la geo–imagen. Para especializar cada objeto geográfico, se propone un algoritmo de especialización que considera la geometría y relaciones topológicas entre los objetos geográficos, tomando como base una ontología de aplicación del dominio geográfico. El conjunto de etiquetas resultante describe la semántica de una geo–imagen.

Palabras clave: Procesamiento de imágenes y visión por computadora, análisis de escena, reconocimiento de objetos.

DESCARGAR ARTÍCULO EN FORMATO PDF

References

1. Adams, N. J., & Williams, C. K. I. (2003). Dynamic trees for image modeling. Image Vision Computing, 21(10), 865 – 877.         [ Links ]

2. Angulo, J., & Serra, J. (2003). Mathematical Morphology in Color Spaces Applied to the Analysis of Cartographic Images. In S. Levachkine, J. Serra & M. Egenhofer (Eds.), Second International Workshop on Semantic Processing of Spatial Data, Mexico City, Mexico, 59–66.         [ Links ]

3. Bezdeck, J.C. (1981). Pattern Recognition with Fuzzy Objective Function Algorithms. Plenum Press, New York.         [ Links ]

4. Borst W. N. (1997). Construction of Engineering Ontologies for Knowledge Sharing and Reuse. Ph.D. Thesis, University of Twente, Enschede, Netherlands.         [ Links ]

5. Bradshaw, B. (2000). Semantic Based Image Retrieval: A Probabilistic Approach, Proceedings of the eighth ACM international conference on Multimedia (MULTIMEDIA'00), Los Angeles, CA, USA, 167 – 176.         [ Links ]

6. Byung–Gyu, K., Jae–Ick, S., & Dong–Jo, P. (2003). Fast image segmentation based on multi–resolution analysis and wavelets. Pattern Recognition Letters, 24 (16), 2995 – 3006.         [ Links ]

7. Junqing Ch., Pappas, T.N., Mojsilovic, A., & Rogowitz, B.E. (2005). Adaptive Perceptual Color–Texture Image Segmentation. IEEE Transactions on Image Processing, 14(10), 1524 – 1536.         [ Links ]

8. Corcho, O., Fernández–López, M., & Gómez–Pérez, A. (2002). Methodologies, tools and languages for building ontologies. Where is the meeting point?. Data & Knowledge Engineering, 46(1), 41 – 64.         [ Links ]

9. Din–Yuen, C., Chih–Hsueh, L., & Wen–Shyong, H. (2005). Image Segmentation with Fast Wavelet–Based Color Segmentation and Directional Region Growing. IEICE Transactions on Information and Systems, E88–D(10), 2249 – 2259.         [ Links ]

10. Fonseca, F., Egenhofer, M., Davis, C., & Câmara, G. (2002). Semantic Granularity in Ontology–Driven Geographic Information Systems. Annals of Mathematics and Artificial Intelligence, 36(1–2), 121 – 151.         [ Links ]

11. Huang J., Kumar S.R., & Zabih R. (1998). An Automatic Hierarchical Image Classification Scheme. Sixth ACM International Conference on Multimedia (MULTIMEDIA '98), Ithaca, NY, USA, 219 – 228.         [ Links ]

12. Levachkine, S. (2003). Raster to Vector Conversion of Color Cartographic Maps. In J. Lladós & Y.B. Kwon (Eds.), Graphics Recognition, Recent Advances and Perspectives, Lecture Notes in Computer Science, 3088, 50 – 62.         [ Links ]

13. Levachkine, S., Velázquez, A., Alexandrov, V., & Kharinov M. (2001). Semantic Analysis and Recognition of Raster–Scanned Color Cartographic Images. In Dorotea Blostein & Young–Bin Kwon (Eds.), 4th International Workshop Graphics Recognition, Algorithms and Applications, Ontario, Canada, 178 – 189.         [ Links ]

14. Jianqing, L; Yee–Hong Y. (1994). Multiresolution Color Image Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 16(7), 689 – 700.         [ Links ]

15. Mueller, M., Segl, & K. Kaufmann, H. (2004). Edge and region–based segmentation technique for the extraction of large, man–made objects in high–resolution satellite imagery. Pattern Recognition, 37(8), 1619 – 1628.         [ Links ]