Power Automate PDF with Embedded Raster Image of Text is not converting using OCR

[Gregory5E99]

I'm using the new PDF to Excel feature and I have mixed raster based images and text inside PDF files.

The raster based image is a table format showing borders with numbers in the table cells.

This is coming over in the excel file as just an image and not OCR operations are occuring.

I don't see any specific settings I can use for how it is converting the PDF file.

Is there a way to force OCR on all raster based images within the PDF?

Thanks!

[Joel Geraci]

You would use the OCR service first and then export to Excel.

[Gregory5E99]

Hello @Joel Geraci ,

I am using the new Power Automate connector https://helpx.adobe.com/document-cloud/help/pdf-connector-for-microsoft-power-automate.html to perform the OCR and convert to excel but I don't see many options like to force OCR on all images.

Is there another way to automate this with Power Automate?

[Joel Geraci]

The OCR is a separate service but if you have a mixture of text and image, I don't think it's going to work.

[Gregory5E99]

Thanks for the suggestion to use OCR first.

Through my testing it seems that PDF to Excel seems to create the same result as PDF to OCR to Excel.

Both processes perform OCR and both have trouble processing OCR on all images withing the PDF.

[Joel Geraci]

I'm fairly certain that the OCR service only works on image-only PDF. I don't think we have a solution to convert a mixture of text and image to just text.

[Gregory5E99]

Thanks Joel,

I am running mixed pdf's and some work but it's hit and miss.

I have a meeting with an Adobe Developer Relation Speciailist Friday.

I'll let you know if anything changes.