Datasets

From TC11
Revision as of 16:02, 16 November 2009 by Masa (talk | contribs) (Machine-print OCR)
Jump to: navigation, search

Optical Character Recognition (OCR)

Machine-print OCR

Handwriting

On-line: vectorial, <math>(x_t, y_t)</math>

  • Kuchibue & Nakayosi (by Masaaki Nakagawa and Stefan Jaeger)
    • Together, these databases comprise more than 3 million Japanese characters from 283 writers.

Off-line: image, <math>I(x,y)</math>

  • IAM Database - A full English sentence database for off-line handwriting recognition.
  • MARG- Medical Article Records Groundtruth ([1]) is a freely-available repository of document page images and their associated textual and layout data. The data has been reviewed and corrected to establish its "ground truth". Please contact Dr. George Thoma (thoma@lhc.nlm.nih.gov) at the National Library of Medicine for more information.

Miscellaneous Kanji handwritten OCR databases

Combined on-line/offline handwriting

  • IRONOFF by the IRESTE group of the University of Nantes, France, is the first database of this type which is available. It combines handwritten samples of words in image (pixel) and in vector format.
Contact: Christian VIARD-GAUDIN (cviard@ireste.fr)