Broken News: Making Newspapers Accessible to Print-Impaired
This addresses accessibility challenges for blind and low-vision people by improving digitization accuracy, though it is incremental as it builds on existing Mask-RCNN and OCR methods.
The paper tackled the problem of making printed newspapers accessible to print-impaired individuals by digitizing them into HTML using layout analysis and OCR, and it reduced the Word Error Rate of news article text by 32.5% with a proposed EdgeMask loss function.
Accessing daily news content still remains a big challenge for people with print-impairment including blind and low-vision due to opacity of printed content and hindrance from online sources. In this paper, we present our approach for digitization of print newspaper into an accessible file format such as HTML. We use an ensemble of instance segmentation and detection framework for newspaper layout analysis and then OCR to recognize text elements such as headline and article text. Additionally, we propose EdgeMask loss function for Mask-RCNN framework to improve segmentation mask boundary and hence accuracy of downstream OCR task. Empirically, we show that our proposed loss function reduces the Word Error Rate (WER) of news article text by 32.5 %.