Tan, Yu (2002) Recognition of ambiguous pairs of totally unconstrained handwritten numerals. Masters thesis, Concordia University.
![[img]](http://spectrum.library.concordia.ca/style/images/fileicons/application_pdf.png)
| PDF 2423Kb |
Abstract
In order to improve the recognition rate of totally unconstrained handwritten numerals, a verification stage is added after classification to distinguish the ambiguous numerals between two or more classes. Two recognition methods, structural method and statistical method, are used to overcome the limitations of a single method and deliver a much more reliable recognition system. The recognition rate is 71.32% using structural method without verification. The verification stage improves the recognition rate from 71.32% to 87.50%. After combining the structural method with statistical method, the final recognition rate reached up to 95.59%
| Divisions: | Concordia University > Faculty of Engineering and Computer Science > Computer Science and Software Engineering |
|---|---|
| Item Type: | Thesis (Masters) |
| Authors: | Tan, Yu |
| Pagination: | xii, 67 leaves : ill. ; 29 cm. |
| Institution: | Concordia University |
| Degree Name: | Theses (M.Comp.Sc.) |
| Program: | Computer Science and Software Engineering |
| Date: | 2002 |
| Thesis Supervisor(s): | Suen, Ching Y |
| ID Code: | 1584 |
| Deposited By: | Concordia University Libraries |
| Deposited On: | 27 Aug 2009 13:20 |
| Last Modified: | 08 Dec 2010 10:21 |
| Related URLs: |
Repository Staff Only: item control page
