Please use this identifier to cite or link to this item:
https://hdl.handle.net/11681/5076| Title: | Feature component reduction through divergence analysis |
| Authors: | Crombie, Michael A. Friend, Nancy J. Rand, Robert S. |
| Keywords: | Ad hoc texture measure Classification Divergence Edge texture measure Feature component reduction Max-min texture measure Relaxation |
| Publisher: | U.S. Army Engineer Topographic Laboratories. Engineer Research and Development Center (U.S.) |
| Series/Report no.: | ETL ; 0305. |
| Description: | Research note Abstract: An approach to developing a reduced set of texture-like descriptors is presented. Results of several experiments using multi-component vectors are presented including experiments with relaxation techniques to cleanse the test images. In view of these results, the practicality of using more complex feature vectors is questioned. |
| Rights: | Approved for public release; distribution is unlimited. |
| URI: | http://hdl.handle.net/11681/5076 |
| Appears in Collections: | Research Note |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| ETL-0305.pdf | 2.69 MB | Adobe PDF |  View/Open |