RNN-LSTM based beta-elliptic model for online handwriting script identification

Recurrent Neural Network (RNN) has achieved the state-of-the-art performance in a wide range of applications dealing with sequential input data. In this context, the proposed system aims to classify the online handwriting scripts based on their labelled pseudo-words. To avoid the vanishing gradient problem, we have used a variant of recurrent network with Long Short-Term Memory. The representation of the sequential aspect of the data is done through the beta-elliptic model. It allows extracting the dynamics and kinematics profiles of different strokes constituting a script over the time. This system was assessed with a large vocabulary containing scripts from ADAB, UNIPEN and PENDIGIT databases. The experiments results show the effectiveness of the proposed system which reached a high recognition rate with only one recurrent layer and using the dropout

[1]  Volker Märgner,et al.  Identification of Arabic/French Handwritten/Printed Words using GMM-Based System , 2014, CORIA-CIFED.

[2]  Monji Kherallah,et al.  Hybrid TDNN-SVM Algorithm for Online Arabic Handwriting Recognition , 2016, HIS.

[3]  A. Alimi.  Beta neuro-fuzzy systems , 2003 .

[4]  Giuseppe Pirlo,et al.  Script Identification of Multi-Script Documents: A Survey , 2017, IEEE Access.

[5]  Christopher Kermorvant,et al.  Dropout Improves Recurrent Neural Networks for Handwriting Recognition , 2013, 2014 14th International Conference on Frontiers in Handwriting Recognition.

[6]  Andrew W. Senior,et al.  Long short-term memory recurrent neural network architectures for large scale acoustic modeling , 2014, INTERSPEECH.

[7]  Hamid Amiri,et al.  Discrimination between Arabic and Latin from bilingual documents , 2011, 2011 International Conference on Communications, Computing and Control Applications (CCCA).

[8]  Anil K. Jain,et al.  Online handwritten script recognition , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[9]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[10]  P. S. Upparamani,et al.  Script Identification from Multilingual Text Documents , 2015 .

[11]  Inchul Song,et al.  RNNDROP: A novel dropout for RNNS in ASR , 2015, 2015 IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU).

[12]  Virach Sornlertlamvanich,et al.  Recurrent Neural Network with Word Embedding for Complaint Classification , 2016, WLSI/OIAF4HLT@COLING.

[13]  Mita Nasipuri,et al.  Word-level script identification for handwritten Indic scripts , 2015, 2015 13th International Conference on Document Analysis and Recognition (ICDAR).

[14]  Geoffrey E. Hinton,et al.  Speech recognition with deep recurrent neural networks , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[15]  Christian Viard-Gaudin,et al.  Information Retrieval Model for Online Handwritten Script Identification , 2009, 2009 10th International Conference on Document Analysis and Recognition.

[16]  Abdel Belaïd,et al.  Co-occurrence Matrix of Oriented Gradients for word script and nature identification , 2015, 2015 13th International Conference on Document Analysis and Recognition (ICDAR).

[17]  J. MacQueen.  Some methods for classification and analysis of multivariate observations , 1967 .

[18]  Yousri Kessentini,et al.  A HMM-Based Arabic/Latin Handwritten/Printed Identification System , 2016, HIS.

[19]  Marcus Liwicki,et al.  A sequence learning approach for multiple script identification , 2015, 2015 13th International Conference on Document Analysis and Recognition (ICDAR).

[20]  Tsuyoshi Murata,et al.  {m , 1934, ACML.

[21]  Monji Kherallah,et al.  Identification and verification system of offline handwritten signature using fractal approach , 2014, International Image Processing, Applications and Systems Conference.

[22]  Adel M. Alimi,et al.  Static and dynamic features for writer identification based on multi-fractals , 2014, Int. Arab J. Inf. Technol..

[23]  Yoshua Bengio,et al.  Investigation of recurrent-neural-network architectures and learning methods for spoken language understanding , 2013, INTERSPEECH.

[24]  Monji Kherallah,et al.  A Time Delay Neural Network for Online Arabic Handwriting Recognition , 2016, ISDA.

[25]  Adel M. Alimi,et al.  On-line Arabic handwriting recognition system based on visual encoding and genetic algorithm , 2009, Eng. Appl. Artif. Intell..

[26]  Amar Mitiche,et al.  On-line handwritten digit recognition based on trajectory and velocity modeling , 2008, Pattern Recognit. Lett..

[27]  Afef Kacem,et al.  Pyramid histogram of oriented gradient for machine-printed/handwritten and Arabic/Latin word discrimination , 2014, 2014 6th International Conference of Soft Computing and Pattern Recognition (SoCPaR).

[28]  Adel M. Alimi,et al.  Online Arabic writer identification based on Beta-elliptic model , 2015, 2015 15th International Conference on Intelligent Systems Design and Applications (ISDA).

[29]  Adel M. Alimi,et al.  Three decision levels strategy for Arabic and Latin texts differentiation in printed and handwritten natures , 2007, Ninth International Conference on Document Analysis and Recognition (ICDAR 2007).

[30]  C. V. Jawahar,et al.  Can RNNs reliably separate script and language at word and line level? , 2015, 2015 13th International Conference on Document Analysis and Recognition (ICDAR).

[31]  Abdel Belaïd,et al.  Identification of Machine-Printed and Handwritten Words in Arabic and Latin Scripts , 2013, ICDAR.