Abstract
The ability to recognize edges of an object is of fundamental importance in computer vision and image processing. In this paper, we present a fast, efficient method of edge extraction of images using directional tracing algorithm. Our work builds on the work of R. Nevatia and K.R. Babu [5], who use linear feature extraction to present algorithms for edge extraction. By employing an intuitive principle that an edge pixel should possess local maximum gradient, but having no more domain information or knowledge, we propose an effective, robust strategy to extract edges that integrates directional tracing and feature extraction. We discuss the computational efficiency of our approach and present a schematic for real time implementation of the algorithm. Application of our algorithm on different images shows excellent results. The main advantages of our technique include: (i) it can be used for a wide variety of images; (ii) it is simple and easy to implement; (iii) it is fast; and (iv) the algorithm is flexible to provide performance control.
Original language | English (US) |
---|---|
Pages (from-to) | 217-234 |
Number of pages | 18 |
Journal | Intelligent Automation and Soft Computing |
Volume | 8 |
Issue number | 3 |
DOIs | |
State | Published - Jan 1 2002 |
Externally published | Yes |
Keywords
- Edge detection
- Edge extraction
- Image segmentation
- Low level vision processing
ASJC Scopus subject areas
- Software
- Theoretical Computer Science
- Computational Theory and Mathematics
- Artificial Intelligence