Hybrid Training Data for Historical Text OCR

Martínek J, Lenc L, Král P, Nikolaou A, Christlein V (2019)

Publication Language: English

Publication Type: Conference contribution

Publication year: 2019

Pages Range: 565-570

Conference Proceedings Title: 2019 International Conference on Document Analysis and Recognition (ICDAR)

Event location: Sydney AU

ISBN: 9781728130149

URI: https://www.researchgate.net/profile/Vincent_Christlein/publication/339025333_Hybrid_Training_Data_for_Historical_Text_OCR/links/5e41bf48a6fdccd9659a17e2/Hybrid-Training-Data-for-Historical-Text-OCR.pdf

DOI: 10.1109/ICDAR.2019.00096


Current optical character recognition (OCR) systems commonly make use of recurrent neural networks (RNN) that process whole text lines. Such systems avoid the task of character segmentation necessary for character-based approaches. A disadvantage of this approach is a need of a large amount of annotated data. This can be solved by sing generated synthetic data instead of costly manually annotated ones. Unfortunately, such data is often not suitable for historical documents particularly for quality reasons. This work presents a hybrid approach for generating annotated data for OCR at a low cost. We first collect a small dataset of isolated characters from historical document images. Then, we generate historical looking text lines from the generated characters. Another contribution lies in the design and implementation of an OCR system based on a convolutional-LSTM network. We first pre-train this system on hybrid data. Afterwards, the network is fine-tuned with real printed text lines. We demonstrate that this training strategy is efficient for obtaining state-of-the-art results. We also show that the score of the proposed system is comparable or even better in comparison to several state-of-the-art systems.

Authors with CRIS profile

Involved external institutions

How to cite


Martínek, J., Lenc, L., Král, P., Nikolaou, A., & Christlein, V. (2019). Hybrid Training Data for Historical Text OCR. In 2019 International Conference on Document Analysis and Recognition (ICDAR) (pp. 565-570). Sydney, AU.


Martínek, Jiří, et al. "Hybrid Training Data for Historical Text OCR." Proceedings of the 2019 International Conference on Document Analysis and Recognition (ICDAR), Sydney 2019. 565-570.

BibTeX: Download