NN/FST - Unsupervised OCR-Postcorrection based on Neural Networks and Finite-state Transducers

• Applicant: Professor Dr. Gerhard Heyer
• Subject Area: Image and Language Processing, Computer Graphics and Visualisation, Human Computer Interaction, Ubiquitous and Wearable Computing ; Data Management, Data-Intensive Systems, Computer Science Methods in Business Informatics ; Software Engineering and Programming Languages
• Term: from 2018 to 2020
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 394341797

The project aims to develop a ready to use software für modul 3 (text optimization) of the OCR-D architecture. The focus of development is in area 3.B (postcorrection) where we plan to also evaluate some up-to-date OCR systems (area 3.A). The main technologies that we plan to use are neural nets (NN) combined with finite-state transducers (FST) to decode recognized lines of text within a noisy-channel model.

• DFG Programme: Research data and software (Scientific Library Services and Information Systems)