Features Extraction and On-line Recognition of Isolated Arabic Characters
| Autor: | Boukelif Aoued, Benbakreti Samir |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Artificial neural network
Computer science Time delay neural network business.industry computer.software_genre Convolutional neural network ComputingMethodologies_PATTERNRECOGNITION Handwriting recognition Noun ComputingMethodologies_DOCUMENTANDTEXTPROCESSING Artificial intelligence Line (text file) business Cursive Arabic script computer Natural language processing |
| Zdroj: | Intelligent Natural Language Processing: Trends and Applications ISBN: 9783319670553 |
| DOI: | 10.1007/978-3-319-67056-0_23 |
| Popis: | In this paper, we present an introduction for the conception of a recognizer for on-line Arabic handwriting. On-line handwriting recognition of Arabic script is a difficult problem, since it is naturally both cursive and unconstrained. This recognizer permits to interpret a script represented by the pen trajectory. This technique is used notably in the PDA electronic diaries. First of all, we will construct a data base with several scripters. Afterwards, and before attacking the recognition phase, there is a constructional phase of samples of the Arabic characters acquired from an electronic tablet to digitize (NOUN DATABASE). This work presents the survey, the implementation and the test for a particular neural network: TDNN (Time Delay Neural Network), applied to the recognition of on-line handwritten writing. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|