Abstract
This paper presents a knowledge-based image segmentation tool for organ delineation in CT (Computed Tomography) images. The noise and low contrast make the detection difficult. Therefore in this method, radial search, noise reduction method and post-processing algorithm have been implemented to improve the quality of contour detection. Three edge detection algorithms have been used and after detection several optimization methods have been employed to get the accurate contour from three detected contours. Finally to achieve higher accuracy of detection, active contour model (ACM), snake, has been used after the contour detected by previous methods.
Original language | English |
---|---|
Title of host publication | Pattern Recognition: 26th DAGM Symposium, Tübingen, Germany, August 30 - September 1, 2004. Proceedings |
Editors | Carl Edward Rasmussen, Heinrich H. Bülthoff, Bernhard Schölkopf, Martin A. Giese |
Publisher | Springer Verlag |
Pages | 504-511 |
Number of pages | 8 |
Volume | Lecture Notes in Computer Science 3175 |
ISBN (Print) | 978-3-540-22945-2 |
DOIs | |
Publication status | Published - 2004 |
Event | DAGM Symposium 2004 - Tubingen, Germany Duration: 30 Aug 2004 → 1 Sept 2004 |
Conference
Conference | DAGM Symposium 2004 |
---|---|
Country/Territory | Germany |
City | Tubingen |
Period | 30/08/04 → 1/09/04 |
Bibliographical note
This article is not yet available on the repository. The paper was given at the 26th DAGM Symposium, Tübingen, Germany, August 30 - September 1, 2004.Keywords
- Pattern Recognition
- Image Processing and Computer Vision
- Artificial Intelligence (incl. Robotics)
- Computer Graphics
- Algorithm Analysis and Problem Complexity