Work

Automatic fuzzy inference system development for marker-based watershed segmentation

Year: 2007
Type: conference-paper
Abstract: Texture image segmentation is a constant challenge in digital image processing. The partition of an image into regions that allow the experienced observer to obtain the necessary information can be done using a Mathematical Morphology tool called the Watershed Transform. This transform is able to distinguish extremely complex objects and is easily adaptable to various kinds of images. The success of the Watershed Transform depends essentially on the existence of unequivocal markers for each of the objects of interest. The standard methods for marker detection are highly specific and complex when objects presenting great variability of shape, size and texture are processed. This paper proposes the automatic generation of a fuzzy inference system for marker detection using object selection done by the expert. This method allows applying the Watershed Transform to biomedical images with diferent kinds of texture. The results allow concluding that the method proposed is an effective tool for the application of the Watershed Transform.
Language: English
Open Access status: diamond
OpenAlex ID: W2023032012
FWCI: 0.274
Cites: 12
Cited by: 2
Related to: 9