Title :
Off-Line Writer Recognition for Farsi Text
Author :
Rafiee, Ali ; Motavalli, Hamidreza
Author_Institution :
Islamic Azad Univ.-Kazeroun Branch, Kazeroun
Abstract :
A new off-line writer recognition method for Farsi text is presented in this paper. 8 different types of features obtained from the handwritten line of text were considered to identify writers based on theirs handwritten. These features are associated with height and width of text. A typical feed forward neural network was used for classification. This method was applied to 20 writers who wrote 5 to 7 lines and 86.5% recognition rate was obtained.
Keywords :
feature extraction; feedforward neural nets; handwriting recognition; pattern classification; text analysis; Farsi text; feed forward neural network; handwritten text line; offline writer recognition; Artificial intelligence; Feature extraction; Feedforward neural networks; Feeds; Handwriting recognition; Histograms; Hydrogen; Neural networks; Text recognition; Writing; Farsi writer; Neural network; Writer identification;
Conference_Titel :
Artificial Intelligence - Special Session, 2007. MICAI 2007. Sixth Mexican International Conference on
Conference_Location :
Aguascallentes
Print_ISBN :
978-0-7695-3124-3
DOI :
10.1109/MICAI.2007.37
