training the model
e-rara: Recognizing mathematical Formulas and Tables
The ETH Library enables access to a large number of scientific titles on e-rara.ch, which are provided with OCR. However, these old prints often also contain mathematical formulas and tables. Such content is largely lost during OCR processing, and often only individual numbers or letters are recognized. Special characters, systems of equations and tabular arrangements are typically missing in the full text.
The aim of this project is to develop a procedure for selected content, how this information could be restored.
Data
Images & Full texts from e-rara.ch
OAI: https://www.e-rara.ch/oai/?verb=Identify
Contact
Team Rare Books and Maps ETH Library
Melanie, Oliver, Sidney, Roman
ruk@library.ethz.ch
Event finished
17.04.2021 17:30
17.04.2021 13:23
~
ruk_ethbib
Event started
16.04.2021 08:30
Joined the team
01.04.2021 13:27
~
ruk_ethbib
Challenge posted
01.04.2021 13:27
~
ruk_ethbib