Implement tesseract video-to-text

[Claiyc]

Issue description

Implement tesseract js methods for live text recognition in the video feed provided by #17
This should already take into account the different rectangles (capture areas) that define where text should be scanned.

See https://maheshlocker.com/building-a-real-time-video-text-recognition-app-using-tesseract-js-and-react for reference.

Tesseract examples with rectangles & multiple workers: https://github.com/naptha/tesseract.js/blob/master/docs/examples.md

• scale capture area width to window(mediastream) width #63

Definition Of Ready (DoR)

This issue can be worked on if

• a developer is assigned
• the issue is labeled
• the issue is assigned to the current sprint
• the issue is estimated
• a meaningful title has been set
• the issue is described in detail

[Claiyc]

Hey team! Please add your planning poker estimate with Zenhub @BeierKevin @FloBaumi @roman533