The Maurdor Project: Improving Automatic Processing of Digital Documents

Brunessaux, Stéphan; Gibert, Patrick; Grilhères, Bruno; Manta, Mathieu; Bodin, Maylis; Choukri, Khalid; Galibert, Olivier; Kahn, Juliette

doi:10.1109/das.2014.58

Cited by 40 publications

(24 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The number of predictions depended on the width of the image. The OM was trained with multiple corpora [11], [12], [13] which included an undocumented database with approximately 153K lines of an historic digitised database 11 and then retrained with the new Bentham data. The system used an hybrid word/character LM that comprised a toplevel LM (3-gram on words/punctuation) for the most frequent words (30k) and a secondary-level LM (10-gram on characters) that dealt with OOV words [14].…”

Section: Competition Protocolmentioning

confidence: 99%

See 1 more Smart Citation

ICFHR2014 Competition on Handwritten Text Recognition on Transcriptorium Datasets (HTRtS)

Sánchez

Romero

Toselli

et al. 2014

2014 14th International Conference on Frontiers in Handwriting Recognition

View full text Add to dashboard Cite

IEEESánchez Peiró, JA.; Romero Gómez, V.; Toselli, AH.; Vidal Ruiz, E. (2014) Abstract-A contest on Handwritten Text Recognition organised in the context of the ICFHR 2014 conference is described. Two tracks with increased freedom on the use of training data were proposed and three research groups participated in these two tracks. The handwritten images for this contest were drawn from an English data set which is currently being considered in the tranScriptorium project. The the goal of this project is to develop innovative, efficient and cost-effective solutions for the transcription of historical handwritten document images, focusing on four languages: English, Spanish, German and Dutch. For the English language, the so-called "Bentham collection" is being considered in tranScriptorium. It encompasses a large set of manuscripts written by the renowned English philosopher and reformer Jeremy Bentham (1748-1832). A small subset of this collection has been chosen for the present HTR competition. The selected subset has been written by several hands (Bentham himself and his secretaries) and entails significant varibilities and difficulties regarding the quality of text images and writting styles. Training and test data were provided in the form of carefully segmented line images, along with the corresponding transcripts. The three participants achieved very good results, with transcription word error rates ranging from 15.0% down to 8.6%.

show abstract

Section: Competition Protocolmentioning

confidence: 99%

“…Hyphenated words were taken as one entire word instead of including its parts. CITlab used Backpropagation-Through-Time (BPTT) [16] using the CTC algorithm [15] for network training 13 .…”

Section: Competition Protocolmentioning

confidence: 99%

ICFHR2014 Competition on Handwritten Text Recognition on Transcriptorium Datasets (HTRtS)

Sánchez

Romero

Toselli

et al. 2014

2014 14th International Conference on Frontiers in Handwriting Recognition

View full text Add to dashboard Cite

show abstract

“…We performed most of the experiments on the handwritten lines in French from the Maurdor dataset [20]. This dataset has the particularity of being very challenging with heterogeneous images from different kind of documents (forms, letters, drawings, ...) and various scanning procedures.…”

Section: Experiments a Experimental Setupmentioning

confidence: 99%

Manifold Mixup Improves Text Recognition with CTC Loss

Moysset¹,

Messina²

2019

2019 International Conference on Document Analysis and Recognition (ICDAR)

View full text Add to dashboard Cite

Modern handwritten text recognition techniques employ deep recurrent neural networks. The use of these techniques is especially efficient when a large amount of annotated data is available for parameter estimation. Data augmentation can be used to enhance the performance of the systems when data is scarce. Manifold Mixup is a modern method of data augmentation that meld two images or the feature maps corresponding to these images and the targets are fused accordingly. We propose to apply the Manifold Mixup to text recognition while adapting it to work with a Connectionist Temporal Classification cost. We show that Manifold Mixup improves text recognition results on various languages and datasets.

show abstract

“…(2) in which G j is the j th gallery image, whose average is λ j , µ i is the mean of those parts of the i th text segment S i underlying G j (x + x , y + y ) and is a function of x and y, is the NCC operator and C i,j is the matching image, whose elements C i,j (x, y) are the matching scores in the range of [0,1]. The location and value of the maximum of C i,j is computed as follows, …”

Section: Parallelised Template Matchingmentioning

confidence: 99%

“…It has numerous applications, particularly in (improving) Optical/Intelligent Character Recognition (OCR/ICR), automatic document analysis and anonymisation [1]. Since the outputs are one of the two classes, binary classification techniques have widely been used to resolve this problem.…”

Section: Introductionmentioning

confidence: 99%