Incorporate image alignment capabilities into OCR backend

CDCgov / ReportVision

Intelligent Data Workflow Automation

Apache License 2.0

1 stars 2 forks source link

Incorporate image alignment capabilities into OCR backend #226

Closed bora-skylight closed 1 week ago

bora-skylight commented 1 month ago

Implement the findings from the image alignment spike into the OCR pipeline

Acceptance Criteria Implement an ImageAligner class in the services module Implement a align function that takes a source_image, template_image, and returns an ndarray aligned image. Please include unit+integration tests

Additional context Alignment code is in the OCR/Alignment folder, consult with @jonchang Alignment Spike findings [1][2]

jonchang commented 3 weeks ago

On hold due to https://github.com/CDCgov/ReportVision/issues/320 being vastly more important

jonchang commented 2 weeks ago

Restarted work on this. Attempting to fix transformation matrix bugs

jonchang commented 1 week ago

Merged as #313, demoed in standup 10/31