Recurrent neural networks (RNNs) with Long Short-Term memory cells currently hold the best known results in unconstrained handwriting recognition. We show that their performance can be greatly improved using dropout - a recently proposed regularization method for deep architectures. While previous works showed that dropout gave superior performance in the context of convolutional networks, it had never been applied to RNNs. In our approach, dropout is carefully used in the network so that it does not affect the recurrent connections, hence the power of RNNs in modeling sequences is preserved. Extensive experiments on a broad range of handwritten databases confirm the effectiveness of dropout on deep architectures even when the network mainly consists of recurrent and shared connections.

/pdf/dropout-improves-recurrent-neural-networks-for-handwriting-2di6lmbs26.pdf

Dropout Improves Recurrent Neural Networks for Handwriting Recognition

In this paper we demonstrate a modified topology for long short-term memory recurrent neural networks that controls the shape of the squashing functions in gating units. We further propose an efficient training framework based on a mini-batch training on sequence level combined with a sequence chunking approach. The framework is evaluated on publicly available data sets containing English and French handwriting by utilizing a GPU based implementation. Speedups of more than 3x are achieved in training recurrent neural network models which outperform state of the art recognition results.

/pdf/fast-and-robust-training-of-recurrent-neural-networks-for-2j76qbcx5r.pdf

Fast and Robust Training of Recurrent Neural Networks for Offline Handwriting Recognition

Given an image of a handwritten word, a CNN is employed to estimate its n-gram frequency profile, which is the set of n-grams contained in the word. Frequencies for unigrams, bigrams and trigrams are estimated for the entire word and for parts of it. Canonical Correlation Analysis is then used to match the estimated profile to the true profiles of all words in a large dictionary. The CNN that is used employs several novelties such as the use of multiple fully connected branches. Applied to all commonly used handwriting recognition benchmarks, our method outperforms, by a very large margin, all existing methods.

/pdf/cnn-n-gram-for-handwritingword-recognition-3vyy6rqsl2.pdf

CNN-N-Gram for HandwritingWord Recognition

A survey of document image word spotting techniques

Intelligent character recognition using fully convolutional neural networks

This study aims at building an efficient word recognition system resulting from the combination of three handwriting recognizers. The main component of this combined system is an HMM-based recognizer which considers dynamic and contextual information for a better modeling of writing units. For modeling the contextual units, a state-tying process based on decision tree clustering is introduced. Decision trees are built according to a set of expert-based questions on how characters are written. Questions are divided into global questions, yielding larger clusters, and precise questions, yielding smaller ones. Such clustering enables us to reduce the total number of models and Gaussians densities by 10. We then apply this modeling to the recognition of handwritten words. Experiments are conducted on three publicly available databases based on Latin or Arabic languages: Rimes, IAM, and OpenHart. The results obtained show that contextual information embedded with dynamic modeling significantly improves recognition.

Dynamic and Contextual Information in HMM Modeling for Handwritten Word Recognition

This paper describes the system for the recognition of French handwriting submitted by A2iA to the competition organized at ICDAR2011 using the Rimes database.
 This system is composed of several recognizers based on three different recognition technologies, combined using a novel combination method.
 A framework multi-word recognition based on weighted finite state transducers is presented, using an explicit word segmentation, a combination of isolated word recognizers and a language model.
 The system was tested both for isolated word recognition and for multi-word line recognition and submitted to the RIMES-ICDAR2011 competition.
 This system outperformed all previously proposed systems on these tasks.

http://www.quaero.org/media/files/bibliographie/menasri_the_a21a_french_drr2012.pdf

The A2iA French handwriting recognition system at the Rimes-ICDAR2011 competition

We present in this paper an HMM-based recognizer for the recognition of unconstrained Arabic handwritten words. 
The recognizer is a context-dependent HMM which considers variable topology and contextual information for a better modeling of writing units.
We propose an algorithm to adapt the topology of each HMM to the character to be modeled.
For modeling the contextual units, a state-tying process based on decision tree clustering is introduced which significantly reduces the number of parameters. 
Decision trees are built according to a set of expert-based questions on how characters are written.
Questions are divided into global questions yielding larger clusters and precise questions yielding smaller ones.
We apply this modeling to the recognition of Arabic handwritten words. 
Experiments conducted on the OpenHaRT2010 database show that variable length topology and contextual information significantly improves the recognition rate.

Variable length and context-dependent HMM letter form models for Arabic handwritten word recognition

HMM-based systems need observation sequences as input. These observations consist of discrete values or vectors extracted from word images or text lines. In this chapter we explore various types of features which are popular for Arabic cursive handwriting recognition. Some of these features are statistical, based on pixel distributions or local directions. Others are structural, based on the presence of loops, ascenders, or descenders. We show how these features can be efficient within HMM-based systems based on sliding windows or grapheme segmentation.

Features for HMM-Based Arabic Handwritten Word Recognition Systems

Cet article presente un systeme a base de HMM pour la reconnaissance hors ligne de mots manuscrits. Afin de prendre en compte les liaisons entre les caracteres d’un meme mot (monographes), leurs modeles sont consideres dependants de leur contexte (trigraphes). Une telle modelisation augmente de maniere considerable le nombre de parametres a calculer, ce qui nous amene a considerer un partage des parametres. Nous effectuons un clustering sur chaque position d’etat, base sur des arbres de decision qui ont l’avantage, en phase de test, de pouvoir associer un modele connu a un trigraphe non appris. Nous avons teste notre systeme sur la base publique Rimes, et ses performances depassent celles d’un systeme a base de monographes et atteignent plus de 80 % de mots correctement reconnus.

Anne-Laure Bianne-Bernard

Papers

Dynamic and Contextual Information in HMM Modeling for Handwritten Word Recognition

The A2iA French handwriting recognition system at the Rimes-ICDAR2011 competition

Variable length and context-dependent HMM letter form models for Arabic handwritten word recognition

Features for HMM-Based Arabic Handwritten Word Recognition Systems

Modélisation de HMM en contexte avec des arbres de décision pour la reconnaissance de mots manuscrits