OCR test for Digitization folks

Tesseract/ocrmypdf on RHEL9

• default repos (including EPEL) only includes the English data package; we would need to install manually from the tessdata_fast repo
• Tesseract does not support reading from a PDF, hence we'd need to either convert to TIFF or run it through ocrmypdf
• ocrmypdf also includes a watched folder

Tesseract+ocrmypdf Examples

A given folder could be designated as an input folder for a given language eg., new-eng,new-deu,new-yid with corresponding output folders, done-eng etc.

One of the challenges for this project is the fact that we have multi-language documents that Abbyy requires us to preselect which languages to recognize, ie., it's an all or nothing proposition.

Example workflow with a custom language

Textract

• Homepage: https://aws.amazon.com/textract/

• code samples: https://github.com/aws-samples/amazon-textract-code-samples

• how would normal end-users drop stuff into S3 Bucket for OCR'ing: https://docs.aws.amazon.com/textract/latest/dg/API_DetectDocumentText.html

Azure