Show simple item record

NCHLT Optical Character Recognition for South African Languages
An OCR system is an application that enables one to convert scanned paper documents into editable and searchable texts. The engine analyses the structure of document image and divides the page into elements such as blocks of texts, tables and images. These blocks are used to identify character image patterns which are used to advance several hypotheses about the character possibilities. These hypotheses are used to produce different character, word and line level variations and associated probabilities. The set of probability hypotheses are then searched to find the most likely combination of characters, words and lines to produce a textual representation of the image.
Roald Eiselen
Roald.Eiselen@nwu.ac.za
North-West University; Centre for Text Technology (CTexT)
Creative Commons Attribution 3.0 Unported License (CC BY 3.0): https://creativecommons.org/licenses/by/3.0/za/
Afrikaans; English; isiNdebele; isiXhosa; isiZulu; Sesotho sa Leboa (Sepedi); Setswana; Sesotho; Siswati; Tshivenda; Xitsonga
Martin Puttkammer; Justin Hocking; Roald Eiselen
Hocking, J. and Puttkammer, M., 2016, November. Optical character recognition for South African languages. In Pattern Recognition Association of South Africa and Robotics and Mechatronics International Conference (PRASA-RobMech), 2016 (pp. 1-5). IEEE.
https://hdl.handle.net/20.500.12185/322
Text
Tools
1.0.
UTF8
NCHLT Text III
Tesseract-OCR
Resource Catalogue
Resource Index
afr; eng; nbl; xho; zul; sot; nso; tsn; ssw; ven; tso
2018-02-05T20:22:45Z; 2018-03-05T17:46:33Z
2018-02-05T20:22:45Z; 2018-03-05T17:46:33Z
2017-02-23


Files in this item

Thumbnail

This item appears in the following Collection(s)

  • Resource Catalogue [232]
    A collection of language resources available for download from the RMA of SADiLaR. The collection mostly consists of resources developed with funding from the Department of Arts and Culture.
  • Resource Index [368]
    A collection of language resource metadata mostly collected during the NHN funded technology audit of 2009, as well as the SADiLaR technology audit of 2018. Not all resources in this collection are available for download.

Show simple item record