Howarth, Peter; Yavlinsky, Alexei; Heesch, Daniel and Rüger, Stefan
(2004).
| DOI (Digital Object Identifier) Link: | http://doi.org/10.1007/11519645_72 |
|---|---|
| Google Scholar: | Look up in Google Scholar |
Abstract
We describe our experiments for the Image CLEF medical retrieval task. Our efforts were focused on the initial visual search. A content-based approach was followed. We used texture, localisation and colour features that have been proven by previous experiments. The images in the collection had specific characteristics. Medical images have a formulaic composition for each modality and anatomic region. We were able to choose features that would perform well in this domain. Tiling a Gabor texture feature to add localisation information proved to be particularly effective. The distances from each feature were combined with equal weighting. This smoothed the performance across the queries. The retrieval results showed that this simple approach was successful, with our system coming third in the automatic retrieval task.
| Item Type: | Conference Item |
|---|---|
| ISSN: | 0302-9743 |
| Academic Unit/Department: | Faculty of Science, Technology, Engineering and Mathematics (STEM) > Knowledge Media Institute (KMi) Faculty of Science, Technology, Engineering and Mathematics (STEM) |
| Item ID: | 11981 |
| Depositing User: | Users 8580 not found. |
| Date Deposited: | 08 Oct 2008 14:18 |
| Last Modified: | 02 Aug 2016 13:18 |
| URI: | http://oro.open.ac.uk/id/eprint/11981 |
| Share this page: |     |
Altmetrics | Scopus Citations |