Abstract
In this paper regulated morphological operations are defined by extending the fitting interpretation of the ordinary morphological operations. The defined operations have a controllable strictness, and so they are less sensitive to noise and small intrusions or protrusions on the boundaries of shapes. The properties of the defined operations are described, and the relations between them and some other non-linear operations are discussed. Given an existing morphological algorithm, it is possible to try and improve the results obtained by it by using the regulated operations instead of the ordinary operations with strictness that may be optimized according to some optimization criteria. Several examples of the proposed approach are presented.
Original language | English |
---|---|
Pages (from-to) | 947-971 |
Number of pages | 25 |
Journal | Pattern Recognition |
Volume | 32 |
Issue number | 6 |
DOIs | |
State | Published - 1 Jan 1999 |
Keywords
- Document analysis
- Fuzzy morphology
- Image processing
- Mathematical morphology
- Regulated morphology
- Soft morphology
ASJC Scopus subject areas
- Software
- Signal Processing
- Computer Vision and Pattern Recognition
- Artificial Intelligence