Скачать или смотреть TintypeText typewriter OCR: Python File 1 "create_rectangles.py" (Instructions video 2)

Tintype¶Text is a tool enabling you to convert scanned typewritten pages (in JPEG image format) into rich text format (RTF) documents, complete with formatting elements such as text alignment, paragraphs, underline, italics, bold and strikethrough.

A neat feature of Tintype¶Text is that the typos (wrong typewritten characters overlaid with a hashtag) automatically get filtered out, and do not appear in the final RTF text.

This video describes how to use the first Python code file in TintypeText entitled "create_rectangles.py" to perform character segmentation in scanned typewritten image files.

You can get my deep learning models for both typewriters on which I developed the code on my Google Drive (2021 Royal Epoch https://drive.google.com/drive/folder... and 1968 Olivetti Underwood Lettera 33 https://drive.google.com/drive/folder..., where the datasets and other useful information to build your own datasets may be found.

Here is the link to the TintypeText github repository:
https://github.com/LPBeaulieu/Typewri...

For an in-depth explanation of all the most common RTF commands and escapes, please consult: https://www.oreilly.com/library/view/...