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

Abstract

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%.

ICFHR2016 Competition on Handwritten Text Recognition on the READ Dataset

“…The challenges stated in this third edition taking into account the experience of the competition in the previous editions [2], [3] and from the experience and requirements in the READ project were the following. Challenge 1.…”

“…HTR techniques for historical documents have been researched in the past for many languages. But the publicly available reference datasets are usually in English [7], [8], [2] or in other very similar languages (from the language modelling point of view), like Spanish [9], [10] or Catalan [11], just to mention a few. In this competition we introduced a new dataset, this time in German.…”

“…This paper describes the third edition of the Handwritten Text Recognition (HTR) competition organized for the International Conference on Frontiers in Handwriting Recognition (ICFHR) 2016 in the framework of the EU TRANSCRIPTORIUM project [1] the first two editions [2], [3], and now in the framework of the EU READ project 1 . As previous editions, the goal of this competition was to bring together researchers for sharing new techniques and ideas on HTR for historical documents.…”

“…The proposed dataset consisted of a subset of documents from the Ratsprotokolle collection 2 composed of minutes of the council meetings held from 1470 to 1805 (about 30, 000 pages), which is used in the READ project. This dataset is written in Early Modern German.…”

General Overview of ImageCLEF at the CLEF 2015 Labs

“…These focused first on benchmarking QbE approaches [22] 4,5 , although lately, QbS approaches have also been considered in the ICDAR'15 [23] 6 , and in the ICFHR '16 7 . Regarding literature about how to deal with hyphenated words, it is worth mentioning the approaches described in [19,27,28].…”