On the Accuracy of CRNNs for Line-Based OCR: A Multi-Parameter Evaluation
This work addresses OCR for historical documents, offering a practical improvement for digitization efforts, though it is incremental as it optimizes existing methods.
The authors tackled the problem of training a high-quality OCR model for difficult historical typefaces on degraded paper, achieving a 0.44% character error rate with only 10,000 lines of training data, outperforming pretrained models trained on much larger datasets.
We investigate how to train a high quality optical character recognition (OCR) model for difficult historical typefaces on degraded paper. Through extensive grid searches, we obtain a neural network architecture and a set of optimal data augmentation settings. We discuss the influence of factors such as binarization, input line height, network width, network depth, and other network training parameters such as dropout. Implementing these findings into a practical model, we are able to obtain a 0.44% character error rate (CER) model from only 10,000 lines of training data, outperforming currently available pretrained models that were trained on more than 20 times the amount of data. We show ablations for all components of our training pipeline, which relies on the open source framework Calamari.