Skip to main content

New Feature - Image Extraction

Feature Request

Currently my python application sends the financial statements to a different system than yours to extract all the images from the statements. Then we scrub out all the images that are not bank checks. The images that are left are then sent to your API for processing. It would be nice to be able to send the statement once, and ask the server to pull the images, clean any based on size criteria, then process the statement and bank check images and return the data. Even if it has to be multiple requests, instead of queued requests, it would be best to be in a single system, and not using multiple systems/companies to get the processing done.

Posted by Sean Ireland ·
Status: Completed3 comments

Log in to comment and vote

Comments3

  • Ianaré Sévi

    Team•

    Jul 10, 2025

    Hi,

    Could your use case be satisfied with something like this?

    1. Send a PDF containing various texts and images, including images of bank checks.

    2. API returns a list of all the bank checks found, as coordinate information.

    3. Using one of the official Mindee libraries, save each bank check as a separate file on disk.

    4. Loop through the extracted images and send them to Mindee for extracting bank check data.

    • Sean Ireland

      •

      Jul 14, 2025

      I was minimizing the sending part. However i found a quite simple way to extract the checks on the local computer and then send them to mindee for processing. thx

      • Ianaré Sévi

        Team•

        Jul 16, 2025

        Great to hear!

        FYI we are in the process of adding image detection to the API.