Special font for finding errors after OCR
The Distributed Proofreaders community has been engaged in scanning books for the Gutenberg project for more than ten years. Hundreds of volunteers help read texts and correct mistakes.
After automatic recognition, errors are most often found in symbols that are similar to each other, like I, l and 1, O and 0, and so on. If you use a regular font like Times, then such errors can be overlooked. Therefore, a special font was created for the Distributed Proofreaders project in which “similar” characters differ as much as possible from each other.
All characters of the font are designed taking into account the numerous tips and wishes of the project participants, voluntary proofreaders. The goal is to most conveniently and effectively identify typical OCR errors.
')
Unfortunately, the Distributed Proofreaders site for historical reasons does not support Unicode, and the community works with a very limited list of languages that are not Cyrillic. Therefore, the DPCustomMono2 font itself is designed only for Latin characters. But the ideas that the font designers have applied can be used to create the Cyrillic character set.
On this page you can compare DPCustomMono2 with various popular fonts.
After automatic recognition, errors are most often found in symbols that are similar to each other, like I, l and 1, O and 0, and so on. If you use a regular font like Times, then such errors can be overlooked. Therefore, a special font was created for the Distributed Proofreaders project in which “similar” characters differ as much as possible from each other.
All characters of the font are designed taking into account the numerous tips and wishes of the project participants, voluntary proofreaders. The goal is to most conveniently and effectively identify typical OCR errors.
')
Unfortunately, the Distributed Proofreaders site for historical reasons does not support Unicode, and the community works with a very limited list of languages that are not Cyrillic. Therefore, the DPCustomMono2 font itself is designed only for Latin characters. But the ideas that the font designers have applied can be used to create the Cyrillic character set.
On this page you can compare DPCustomMono2 with various popular fonts.
Source: https://habr.com/ru/post/154385/
All Articles