NLP group

Hybrid Training Data for Historical Text OCR

Jiří Martínek and Ladislav Lenc and Pavel Král and Anguelos Nicolaou and Vincent Christlein
15th International Conference on Document Analysis and Recognition (ICDAR 2019) (2019)

PDF

Abstract

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 using 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 Unfortunately, such data is often not suitable for historical documents particularly for quality reasons.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 We first collect a small dataset of isolated characters from historical document images.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.