Thesis (PDF) - Signal & Image Processing Lab
Thesis (PDF) - Signal & Image Processing Lab
Thesis (PDF) - Signal & Image Processing Lab
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
Abstract<br />
This research addresses a new image filtering methodology, based on mathematical<br />
morphology.<br />
In this thesis a new general framework for producing morphological, self-dual<br />
operators that are compatible to a given tree representation is proposed.<br />
For every tree representation, a set of morphological operators on a complete<br />
inf-semilattice in the corresponding tree-representation domain is derived. Morpho-<br />
logical erosion, opening, and opening by reconstruction operators are defined using<br />
this framework. The proposed image filtering scheme consists of three steps:<br />
1. Transform the input image to the corresponding tree representation,<br />
2. perform morphological operators in the tree representation domain, and<br />
3. transform the resulting tree representation back to the image domain.<br />
A particular case of this general framework is presented and studied. It involves a<br />
new tree representation,which we also developed in this research, called the Extrema-<br />
Watershed Tree. The particular case example emphasizes the ability of the general<br />
framework to generate new and useful sets of morphological operators.<br />
A number of potential applications for the Extrema-Watershed Tree is proposed.<br />
The new morphological operators excel in tasks suited for the application of classical<br />
morphological operators, but that require, in addition, self-duality. The proposed<br />
applications are pre-processing for OCR (Optical Character Recognition) algorithms,<br />
de-noising of images, and preprocessing for dust and scratch detection. In addition<br />
we show that the tree has an implicit segmentation property that could be used in<br />
image segmentation algorithms.<br />
In a previous work, Keshet has defined a complete inf-semilattice of images of<br />
alternating sequences. In this research an efficient implementation of morphological<br />
operators in this semilattice is proposed. In addition, the “trench” problem that<br />
1