Related Experiment Videos
Automated image segmentation by fractal grey tone functions
1Unité de Recherches Biomathématiques et Biostatistiques (U.263 I.N.S.E.R.M.), Université Paris 7, France.
Gegenbaurs Morphologisches Jahrbuch
|January 1, 1989
Summary
This study introduces a novel image segmentation technique. The method effectively isolates structures using fractal geometry and mathematical morphology, requiring only basic image processing operations.
Area of Science:
- Image analysis
- Computer vision
- Fractal geometry
Background:
- Image segmentation is crucial for analyzing complex structures.
- Traditional methods struggle with characterizing structures across resolutions.
- Fractal geometry offers a framework for describing irregular shapes.
Purpose of the Study:
- To present a new image segmentation method.
- To leverage mathematical morphology and fractal geometry for structure isolation.
- To enable segmentation using only dilated and eroded grey tone images.
Main Methods:
- The method is derived from mathematical morphology and fractal geometry.
- It utilizes the fractal grey tone function and fractal dimension.
- Image segmentation is achieved through a series of dilated and eroded grey tone images.
Main Results:
- The proposed method successfully isolates structures characterized by fractal properties.
- It demonstrates effectiveness across a range of resolutions.
- The technique requires minimal input, relying on basic image operations.
Conclusions:
- A novel and efficient image segmentation method based on fractal geometry and mathematical morphology has been developed.
- The method provides a robust approach for isolating complex structures.
- Its simplicity in terms of input requirements makes it broadly applicable.