Online and Offline Vocabulary Correction while using a Digital Pen
| Název práce v češtině: | |
|---|---|
| Název v anglickém jazyce: | Online and Offline Vocabulary Correction while using a Digital Pen |
| Akademický rok vypsání: | 2008/2009 |
| Typ práce: | diplomová práce |
| Jazyk práce: | angličtina |
| Ústav: | Ústav formální a aplikované lingvistiky (32-UFAL) |
| Vedoucí / školitel: | prof. Ing. Václav Hlaváč, CSc. |
| Řešitel: | skrytý - zadáno a potvrzeno stud. odd. |
| Datum přihlášení: | 14.11.2008 |
| Datum zadání: | 15.01.2009 |
| Datum a čas obhajoby: | 26.10.2009 00:00 |
| Datum odevzdání elektronické podoby: | 26.10.2009 |
| Datum proběhlé obhajoby: | 26.10.2009 |
| Oponenti: | RNDr. Daniel Průša, Ph.D. |
| Zásady pro vypracování |
| The Actimage Company wants to improve its recognition software by offering a new method for writing correction. It looks for a new technology allowing the writer to come back on his writing in order to correct one or more letters under the conditions that 1) this correction does not affect the recognition by the OCR and 2) the writing visibility remains complete.
In the past, a graphical symbol allowing to the system to realize the specified correction was proposed for each correction rule. This work was limited to the cognitive aspect but not implemented and experimented. The idea of the thesis is to examine again this nomenclature of correction signs, and to propose a methodology for their recognition and to experiment them by the Actimage platform, Actinode coupled with the integrated OCR of Vision Objects. The work steps are: 1) the exam of the graphic signs and the proposition of a definitive lists of signs for Actimage, 2) the integration of some of them in the system to test their recognition, 3) for complex signs, the proposition of new methods for graphical symbol modeling and recognition. In the both last cases, the student has to show how to replace the erroneous letter by the new one in order to allow the OCR working on it. |
| Seznam odborné literatury |
| Sonka M., Hlavac, V., Boyle R.: Image Processing, Understanding, and Machine Vision, 3rd edition, Thomson, Toronto 2007, pp. 829
Pittman, James. A. "Handwriting Recognition: Tablet PC Text input.", IEEE Computer, September 2007, Vol 40 Number 9 pp. 49..54 LiPi Toolkit http://lipitk.sourceforge.net/ |
| Předběžná náplň práce |
| The Actimage Company wants to improve her recognition software by offering a new method for writing correction. She looks for a new technology allowing the writer to come back on his writing in order to correct one or more letters under the conditions that 1) this correction does not affect the recognition by the OCR and 2) the writing visibility remains complete.
In the past, a graphical symbol allowing to the system to realize the specified correction was proposed for each correction rule. This work was limited to the cognitive aspect but not implemented and experimented. The idea of the thesis is to examine again this nomenclature of correction signs, and to propose a methodology for their recognition and to experiment them by the Actimage platform, Actinode coupled with the integrated OCR of Vision Objects. |
| Předběžná náplň práce v anglickém jazyce |
| The Actimage Company wants to improve her recognition software by offering a new method for writing correction. She looks for a new technology allowing the writer to come back on his writing in order to correct one or more letters under the conditions that 1) this correction does not affect the recognition by the OCR and 2) the writing visibility remains complete.
In the past, a graphical symbol allowing to the system to realize the specified correction was proposed for each correction rule. This work was limited to the cognitive aspect but not implemented and experimented. The idea of the thesis is to examine again this nomenclature of correction signs, and to propose a methodology for their recognition and to experiment them by the Actimage platform, Actinode coupled with the integrated OCR of Vision Objects. |