Combination of local and global vision modelling for Arabic handwritten words recognition
Tóm tắt
We propose an Arabic handwritten word recognition system based on the idea of the PERCEPTRO system developed by Cote (Cote et al. (1998)) for Latin word recognition. It is a specific neural network, named transparent neural network, combining a global and a local vision modeling (GVM-LVM) of the word. In the forward propagation movement, the former (GVM) proposes a list of structural features characterizing the presence of some letters in the word. GVM proposes a list of possible letters and words containing these characteristics. Then, in the backpropagation movement, these letters are confirmed or not according to their proximity with corresponding printed letters. The correspondence between the letter shapes and the corresponding printed letters is performed by LVM using the correspondence of their Fourier descriptors, playing the role of a letter shape normalizer.
Từ khóa
#Handwriting recognition #Neural networks #Humans #Shape #Robustness #Feature extraction #Laboratories #Signal processing #Office automation #BankingTài liệu tham khảo
al-badr, 1994, Survey and bibliography of arabic optical text recognition, Signal Processing, 49
10.1109/34.56206
maddouri snoussi, 1999, Une me?thode de reconnaissance de mots manuscrits arabes par re?seaux de neurones transparent, Confe?rence Internationale en Informatique
10.1016/S0031-3203(97)00084-8
10.1142/9789812797681_0003
guillevic, 1995, Unconstrained handwriting recognition applied to the processing of bank checks, A Thesis of the Department of Computer Science
co?te?, 1998, Automatic reading of cursive scripts using a reading model and perceptual concepts, IJDAR, 1, 3, 10.1007/s100320050002
10.1109/34.506792
10.1037//0033-295X.88.5.375
10.1016/0146-664X(82)90034-X
maddouri, 2000, Local normalization towards global recognition of arabic handwritten script, IAPR Workshop Document Analysis Systems
maddouri, 2001, Normalization of handwriting pieces of arabic word: Two different approaches, Tunisian-German Conference on Smart Systems and Devices