Data Extraction From PDFs Containing OCR image.

[Mayank_2611]

I have a PDF , in Page 2 there is a table containing an image …I want to get that table as output in excel/Gsheet. Can Somebody help me here? I am trying using Image Reader and Tess4J nodes but not making any progress.

[ScottF]

Hi @Mayank_2611 and welcome to the forum.

This is a question that comes up now and then. Take a look a this post, maybe it can work as a starting point:

Extract information from a multiple pdfs PDF file

Hi @Sbhandary - Data extraction from PDFs can definitely be a tricky task. Let me point you to some other threads where this type of analysis is address in more detail. The Data Connect recording that Victor links in the second post might be of particular interest.

I went ahead and removed the PDF example file you uploaded since it contains personal information (names and addresses).

[PBJ]

First step with two ways:

1. I’m using Tikka node with embedded pictures saved to a particular folder (not running with the lastest versions of PDF with new format of embedded pictures encoding).
2. I’m using a external open source and free library (poppler) to convert PDF pages into pictures saved to a particular folder.

From this folder, I create images and use Tess4J to make the OCR of pictures.

KNIME_OCR.knwf (66,1 Ko)

For new encoded embedded pictures, I search external free and open source tools.
It should be great if KNIME can create a node