Convert PDF to Images format

MindStudio How-To Guides
1

I want to analyze a PDF file as a set of images, rather than extract text. I’m trying to use the ‘Convert PDF to Images’ block, but I can’t tell what the output variable format is. It says an array of images, but are these URLs, or base64 strings, etc. The ‘quick help’ is completely empty for this block, so I don’t know how I can use this variable.

Most importantly, how can I use this output in the ‘Analyze Image’ block? That block asks for a URL for each image. Can I simply use the output variable from the previous step (Convert PDF to Images)?

It would be incredibly helpful (to many people, I’m sure) if there was an ‘Analyze PDF as Images’ block, which could take a PDF URL and pass it as a set of images to a model for analysis.

2

Hi @mikehochleutner,

Thanks for the post!

The Convert PDF to Images block outputs an array of URLs.

The Analyze Image block takes one URL at a time, so you can’t pass the full array to it directly, but you can loop through each URL using a Run Workflow block that processes each image in a sub-workflow. Here’s an example you can explore:
https://app.mindstudio.ai/agents/pdf-to-images-c3f41e49/remix

3

Thanks for the response. I figured it out, and it’s working great!

I’ve trained it to analyze undergraduate transcripts for MBA applicants, and it does much better than when it was simply parsing text. MBAI: Admissions Intelligence | Homepage