Computerized detection of abnormalities in endoscopic oesophageal images
The research work comprises four segments: extracting lumen boundary, detecting lumen related abnormalities, forming a suitable colour segmentation framework, and classifying abnormalities using colour characteristics, applied to oesophageal images. Considering that an accurate determination of lume...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Theses and Dissertations |
| Published: |
2008
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/3812 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Summary: | The research work comprises four segments: extracting lumen boundary, detecting lumen related abnormalities, forming a suitable colour segmentation framework, and classifying abnormalities using colour characteristics, applied to oesophageal images. Considering that an accurate determination of lumen boundary is important in the navigation of the endoscope in oesophagoscopy, a new method is proposed for automatic detection and extraction of lumen boundary of oesophagus. The proposed technique involves pre-processing, histogram analysis, region growing, post-processing, |
|---|