Offline grammar-based recognition of handwritten sentences

Zimmermann, Matthias; Chappelier, Jean-Cédric; Bunke, Horst

doi:10.1109/tpami.2006.103

Cited by 45 publications

(10 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Note however that the RNN is less affected by the dictionary size, and that for the 30,000 word dictionary, performance continues to increase for the RNN but drops for the HMM. The tendency of HMMs to lose accuracy for very large handwriting lexicons has been previously observed [51].…”

Section: F Influence Of Dictionary Sizementioning

confidence: 69%

“…The 'standard' dictionary we used for our main results consisted of the 20,000 most frequently occurring words in the three corpora. The figure 20,000 was chosen because it had been previously shown to give best results for HMMs [51]. Note that this dictionary was 'open', in the sense that it did not contain all the words in either the online or offline test set.…”

Section: B Language Model and Dictionariesmentioning

confidence: 99%

“…One linear HMM was used for each character. For the online data every character model contained eight states, while for the offline data, the number of states was chosen individually for each character [51]. The observation probabilities were modelled by mixtures of diagonal Gaussians.…”

Section: Hmm Parametersmentioning

confidence: 99%

See 2 more Smart Citations

A Novel Connectionist System for Unconstrained Handwriting Recognition

Graves¹,

Liwicki

Fernández

et al. 2009

IEEE Trans. Pattern Anal. Mach. Intell.

Self Cite

1,709

887

View full text Add to dashboard Cite

Abstract-Recognising lines of unconstrained handwritten text is a challenging task. The difficulty of segmenting cursive or overlapping characters, combined with the need to exploit surrounding context, has led to low recognition rates for even the best current recognisers. Most recent progress in the field has been made either through improved preprocessing, or through advances in language modelling. Relatively little work has been done on the basic recognition algorithms. Indeed, most systems rely on the same hidden Markov models that have been used for decades in speech and handwriting recognition, despite their well-known shortcomings. This paper proposes an alternative approach based on a novel type of recurrent neural network, specifically designed for sequence labelling tasks where the data is hard to segment and contains long range, bidirectional interdependencies. In experiments on two large unconstrained handwriting databases, our approach achieves word recognition accuracies of 79.7% on online data and 74.1% on offline data, significantly outperforming a state-of-the-art HMM-based system. In addition, we demonstrate the network's robustness to lexicon size, measure the individual influence of its hidden layers, and analyse its use of context. Lastly we provide an in depth discussion of the differences between the network and HMMs, suggesting reasons for the network's superior performance.

show abstract

Section: F Influence Of Dictionary Sizementioning

confidence: 69%

Section: B Language Model and Dictionariesmentioning

confidence: 99%

See 1 more Smart Citation

A Novel Connectionist System for Unconstrained Handwriting Recognition

Graves¹,

Liwicki

Fernández

et al. 2009

IEEE Trans. Pattern Anal. Mach. Intell.

Self Cite

1,709

887

View full text Add to dashboard Cite

show abstract

“…Achievement of High reliability was by using a rejection rule. M. Zimmermann et.al in their IEEE transaction paper on pattern analysis and machine intelligence described the working and implementation of an offline system using grammar based recognition of handwritten sentences [5]. R. Plamondon et.al in their IEEE article of pattern analysis and machine intelligence contributed to M. Zimmermann's offline grammar based handwriting recognition system and wrote a comprehensive survey on online and offline handwriting recognition systems which also used optical concepts and variations of supervised learning algorithms [6].…”

Section: Theoretical Backround 21 Related Work In Image Processing Andmentioning

confidence: 99%

A Survey on Optical Handwriting Recognition System using Machine Learning Algorithms

Kumbhar¹,

Kunjir²

2017

IJCA

View full text Add to dashboard Cite

The recent advancement and development in technical fields and logistics has tackled the challenges of science and have delivered a good impact ratio by avoiding issues faced by fully or partially challenged and impaired people. The purpose of this paper is to provide easy facilities and improved services to a wide range of customers from school going children to teachers, professors, home tutors, visually impaired, and literary scholars. The title says it all, the concepts and theoretical background in the paper explains how the handwriting scripted in the image can be converted into textual information using several machine learning algorithms which follow the principles of supervised and unsupervised learning respectively. With the help of microcontrollers, feature extraction material, trainers and image processing concepts the proposed idea can be implemented feasibly. Machine learning algorithms such as Support Vector Machine, Random forest and neural nets algorithms have been compared and evaluated on basis of precision, recall, accuracy and other performance parameters. On comparing the mentioning algorithms it has been observed that 'SVM' algorithm outclasses the comparison test and gives out the best result in terms of accuracy, latency and robustness. On observing the overall effects of algorithms and conversions between handwritten text into textual information (maybe pdf's). It has been concluded that the work can be further implemented on platforms such as e-learning, in which the application can be deployed on internet for android and IoS users with the help of mobile computing principles. Students in the classroom can just click a picture of what's written on the board by the professor and the application will convert the text written on the image into a PDF and save it in the internal memory of your gadget.

show abstract

“…In other recognition systems, the use of several language models during the recognition step can be viewed as a correction step on a baseline recognition system with only one LM. Thus, in [29,32], a n-class LM is combined with another n-class model and with a n-gram model, respectively, whereas in [45], a stochastic context-free grammar is combined with a n-gram model: in both cases, the recognition system benefits from the use of the added language model. Finally, an original correction approach has been proposed in [12].…”

Section: Handwriting Recognitionmentioning

confidence: 99%

Error handling approach using characterization and correction steps for handwritten document analysis

Quiniou

Cheriet

Anquetil

2011

IJDAR

View full text Add to dashboard Cite

In this paper, we present a framework to handle recognition errors from a N -best list of output phrases given by a handwriting recognition system, with the aim to use the resulting phrases as inputs to a higher-level application. The framework can be decomposed into four main steps: phrase alignment, detection, characterization, and correction of word error hypotheses. First, the N -best phrases are aligned to the top-list phrase, and word posterior probabilities are computed and used as confidence indices to detect word error hypotheses on this top-list phrase (in comparison with a learned threshold). Then, the errors are characterized into predefined types, using the word posterior probabilities of the top-list phrase and other features to feed a trained SVM. Finally, the final output phrase is retrieved, thanks to a correction step that used the characterized error hypotheses and a designed word-to-class backoff language model. First experiments were conducted on the ImadocSen-OnDB handwritten sentence database and on the IAM-OnDB handwritten text database, using two recognizers. We present first results on an implementation of the proposed framework for handling recognition errors on transcripts of handwritten phrases provided by recognition systems.

show abstract

Offline grammar-based recognition of handwritten sentences

Cited by 45 publications

References 28 publications

A Novel Connectionist System for Unconstrained Handwriting Recognition

A Novel Connectionist System for Unconstrained Handwriting Recognition

A Survey on Optical Handwriting Recognition System using Machine Learning Algorithms

Error handling approach using characterization and correction steps for handwritten document analysis

Contact Info

Product

Resources

About