Hoekstra, E. (2003) Grayscale Feature Extraction. Master's Thesis / Essay, Computing Science.
|
Text
Infor_Ma_2003_EHoekstra.CV.pdf - Published Version Download (2MB) | Preview |
Abstract
This research examines the effects of grayscale feature extraction on, primary, the recognition of characters and, secondary, the recognition of license plates. Examination of the influence of thresholding on the recognition of characters displays the possible advantage of grayscale feature extraction. Two methods are selected to do this job, pseudo Zernike moments and topographic labeling of pixels. Features from zoning topographic labels show the best result in use with neural networks. The robustness of topographic features is observed when compared to classification with binary features. In the context of license plate recognition this means incorrect classification can be reduced while preserving correct classifications.
| Item Type: | Thesis (Master's Thesis / Essay) |
|---|---|
| Degree programme: | Computing Science |
| Thesis type: | Master's Thesis / Essay |
| Language: | English |
| Date Deposited: | 15 Feb 2018 07:29 |
| Last Modified: | 15 Feb 2018 07:29 |
| URI: | https://fse.studenttheses.ub.rug.nl/id/eprint/8866 |
Actions (login required)
 |
View Item |