Scanned PDF forms/emails without structure - Githubissues

kartAI / ntnu-kpro-ai-assistant

Kundestyrt prosjekt (TDT4290) prosjekt med NTNU studenter. Dette repoet er laget av og for studentene ved NTNU

0 stars 0 forks source link

Scanned PDF forms/emails without structure #48

Closed SverreNystad closed 2 weeks ago

SverreNystad commented 3 weeks ago

The system shall support scanned PDF forms or emails that do not follow a predefined structure. These unstructured documents may require additional OCR processing to extract relevant information.

Acceptance Criteria

[x] The system shall accept unstructured scanned PDF files and emails.
[x] The system should be able to use OCR to extract text from these documents.
[x] Handle potential variations in format and content, ensuring key information is extracted for processing.

Notes

This feature is designed to handle irregular or unstructured document layouts, which may require more advanced OCR processing techniques.

SverreNystad commented 3 weeks ago

@maurice-wegerif look at this: https://pypi.org/project/pytesseract/

SverreNystad commented 2 weeks ago

This is now done