Database to allow search of scanned documents.
Operate on Pareto principle (20% effort, 80% results). A detailed discussion of the app can be found HERE.

Accessing

accessing_ocr_database.mp4

Searching

searching_viewing_downloading.mp4

Structure

• Frontend:

  • Submit scans
  • Upload scans
  • Provide metadata
    • Document Types
  • Search scans
    • By metadata
  • By content (OCR)
  • View scans
    • Path to saved scan

• Backend:

  • Accept scans
  • Store scans
  • Extract metadata
  • Extract text from scans
  • Search scans
    • By metadata
    • By content
  • Return path to saved scan

Notes: - To fix opencv issue, https://docs.streamlit.io/knowledge-base/dependencies/libgl