|
| << vorige | volgende >> |
Tijdschrift beschrijving
Alle jaargangen van het bijbehorende tijdschrift
Alle afleveringen van het bijbehorende jaargang
Alle artikelen van de bijbehorende aflevering
Details van artikel 43 van 49 gevonden artikelen
| Titel: |
Segmentation and Recognition of Handwritten Numeric Chains |
| Auteur: |
Salim Ouchtati Mouldi Bedda Abderrazak Lachouri |
| Verschenen in: |
Journal of computer science |
| Paginering: | Jaargang 3 (2007) nr. 4 pagina's 242-248 |
| Jaar: |
2007 |
| Inhoud: |
Automatic reading of numeric chains has been attempted in several application areas suchas bank cheque processing, postal code recognition and form processing. Such applications have beenvery popular in handwriting recognition research, due to the possibility to reduce considerably themanual effort involved in these tasks. In this study we propose an off line system for the recognition ofthe handwritten numeric chains. Firstly, study was based mainly on the evaluation of neural networkperformances, trained with the gradient back propagation algorithm. Used parameters to form the inputvector of the neural network are extracted on the binary images of the digits by several methods:distribution sequence, Barr features and centred moments of different projections and profiles.Secondly, study was extented for the reading of the handwritten numeric chains constituted of avariable number of digits. Vertical projection was used to segment the numeric chain at isolated digitsand every digit (or segment) was presented separately to the entry of the system achieved in the firstpart (recognition system of the isolated handwritten digits). The performances of the proposed systemfor the used database attain a recognition rate equal to 91.3%. |
| Uitgever: |
Science Publications (provided by DOAJ) |
| Bronbestand: |
Elektronische Wetenschappelijke Tijdschriften |
Details van artikel 43 van 49 gevonden artikelen
| << vorige | volgende >> |