Fuzzy classification improvement by a pre-perceptual labelled segmentation algorithm
The goal of this paper is to present how two different image processing approaches can be enhanced by merging both methodologies. We will see how the results of a perceptual labelled segmentation methodology [7] can be improved by applying a fuzzy classification algorithm [2] based on a fuzzy outran...
| Autores: | , , , |
|---|---|
| Tipo de recurso: | capítulo de libro |
| Fecha de publicación: | 2004 |
| País: | España |
| Institución: | Universidad Complutense de Madrid (UCM) |
| Repositorio: | Docta Complutense |
| Idioma: | inglés |
| OAI Identifier: | oai:docta.ucm.es:20.500.14352/53167 |
| Acceso en línea: | https://hdl.handle.net/20.500.14352/53167 |
| Access Level: | acceso abierto |
| Palabra clave: | 004.8 Computer Science Artificial Intelligence Information Systems Inteligencia artificial (Informática) 1203.04 Inteligencia Artificial |
| Sumario: | The goal of this paper is to present how two different image processing approaches can be enhanced by merging both methodologies. We will see how the results of a perceptual labelled segmentation methodology [7] can be improved by applying a fuzzy classification algorithm [2] based on a fuzzy outranking methodology [9] as a postprocessing algorithm, and viceversa. A comparison of the individual algorithms with the combination of both algorithms will be presented in order to demonstrate the improvement. Color Bone Marrow (1) images will be used. The objective is to detect White Blood Cells. The detection of white blood cells in bone marrow microscopic images presents big difficulties because of the great variance in their characteristics and also because of staining and illumination inconsistences. On the other hand, the maturity classes of white blood cells actually represents a continuum; cells frequently overlap each other, and there is a fairly wide variation in size and shape of nucleus and cytoplasm regions within given cell classes. |
|---|