Research work · 2007

A scale-space toggle operator for morphological segmentation

Leyza Baldo Dorini, Neucimar J. Leite

Biblioteca Digital da Memória Científica do INPE (National Institute for Space Research)

Abstract

The analysis of different representation levels has been largely used to handle the multiscale nature of image data. Here, we explore the scale-space properties of a toggle operator defined on a scaled morphological framework. These properties conduce to a well-controlled image extrema simplification, yielding sound segmentation and filtering results even when the operator is used in a binarization process. Also, the watershed transform using markers extracted from the processed image gives better segmentation results than when using markers from the original one. To show the robustness of our approach, we carried out tests on images of different classes and subjected to different lighting conditions.

Download PDF