Resumo

KOTZE, Gideon  e  WOLFF, Friedel. Exchanging image processing and OCR components in a Setswana digitisation pipeline. SACJ [online]. 2020, vol.32, n.2, pp.218-231. ISSN 2313-7835.  http://dx.doi.org/10.18489/sacj.v32i2.707.

As more natural language processing (NLP) applications benefit from neural network based approaches, it makes sense to re-evaluate existing work in NLP. A complete pipeline for digitisation includes several components handling the material in sequence. Image processing after scanning the document has been shown to be an important factor in final quality. Here we compare two different approaches for visually enhancing documents before Optical Character Recognition (OCR), (1) a combination of ImageMagick and Unpaper and (2) OCRopus. We also compare Calamari, a new line-based OCR package using neural networks, with the well-known Tesseract 3 as the OCR component. Our evaluation on a set of Setswana documents reveals that the combination of ImageMa-gick/Unpaper and Calamari improves on a current baseline based on Tesseract 3 and ImageMagick/Unpaper with over 30%, achieving a mean character error rate of 1.69 across all combined test data.CATEGORIES: • Applied computing ~ Optical character recognition • Computing methodologies ~ Image processing

Palavras-chave : digitisation; optical character recognition; image processing; neural networks.

        · texto em Inglês     · Inglês ( pdf )