The Altair Community is migrating to a new platform to provide a better experience for you. In preparation for the migration, the Altair Community is on read-only mode from October 28 - November 6, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here

How to do OCR with images

alsaqer002alsaqer002 Member Posts: 5 Contributor II
edited November 2018 in Help
Hello all,

I want to do OCR on images in order to get text.

I already installed the (image Miner extension, Text processing extension, Aylien Text Analysis extension, and Feature Selection extension)

after I use the "Open Color Image" or "Open Gray-scale Image" operators, which operator should I use to recognize and extract the text and features from image file?

Any help will be appreciated.

Thanks,

Answers

  • ljlj Member Posts: 10 Contributor II

    Hi. I'm new in here and maybe lucky to ask my question in the right place?

    So what' s the matter? I have a highly aggregated pdf-file containing images, tables and metadata as text. Doing it manually, I would do some save as tables with the selectes tables within the pdf-file, save as image with the selected images and save as csv on the rest. Everything with the Adobe ...

    After some cleaning i would have three seperated files: xlsx, csv and jpeg. The first two can be read in natively in RM the third, i suppose an extension working fine with it.

    How can i do it better? The mentioned elemets appear always at the same position. Horthonworks and Zapier could be integrated to the process.

    I'm looking forward to your contributions. Thank you in advance.

    Lukas

     

     

  • ljlj Member Posts: 10 Contributor II

    Test_PREP_Seite_1.pngTest_PREP_Seite_2.png

    The mentioned file looks like the pictures above.

     

  • VaclavVaclav RapidMiner Certified Expert, Member Posts: 23 Maven

     Hello,

    you need OCR for this, but even if you are able to convert image to text, it will be big problem with structure of the text and extract specific values. You can use Tesseract for OCR. If you are satified with output, you can order custom development of extension for RapidMiner.

     

    Best wishes

    Vaclav

  • ljlj Member Posts: 10 Contributor II

    Hi,

    I somehow didn't notice Your reply. Sorry. And then, as always the time.

    I'll take a look on Tesseract.

    Thank You.

    Beste Whishes

    Lukas

  • Thomas_OttThomas_Ott RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,761 Unicorn

    There is a new extension on the Marketplace that RM just developed called PDF Table Extraction. Maybe that can help?

     

    https://marketplace.rapidminer.com/UpdateServer/faces/product_details.xhtml?productId=rmx_pdf_table_extraction

     

     

  • ljlj Member Posts: 10 Contributor II

    Hi Thomas,

    Wow, that sounds good. I'm excited.Thank You a lot.

    I will leave a feedback after having tried it out.

    Best whishes.

    Lukas

     

     

Sign In or Register to comment.