2

How effective is Adobe ifilter for extracting text from scan\image in a pdf? does it give 100% result or it can have its own limitations.

I am asking this because we have migrated a huge set of documents from Documentum to SharePoint and it shows less number of results here, for example if it showed 200 pdfs in Documentum for a keyword XYZ, it shows 150 in SharePoint for the same keyword.

When we checked missing documents. They are present in SharePoint and crawled as well.

Improve this question
  • Is it possible security on the PDFs hid them from you? The search results are only going to show you what you have access to. – Eric Alexander Jun 2 '15 at 1:13
  • I am using farm admin account, so I guess this should not be the case – SPAdmin Jun 3 '15 at 21:31
0

If the PDF file contains images instead of text, i.e. you cannot select the text using a Acrobat Viewer - SharePoint will not be able to index it even with iFilter for Adobe configured. http://blogs.msdn.com/b/priyo/archive/2006/03/02/542208.aspx . He also mentioned the workaround, i.e convert the PDF into Tiff or using the PDF write to convert the image into text.

  1. That is 1st reason why you seeing the less number
  2. Search results are security trimed, it will show the result from all over where you have access.
Improve this answer
  • The PDFs are already in OCR format still SP has difficulty in searching them. And I am using a farm admin account, so I guess security should not be the issue. – SPAdmin Jun 3 '15 at 21:40
  • Do you have idler installed n full crawl? Which I filter u are using? – Waqas Sarwar MVP♦ Jun 3 '15 at 21:59
  • we have done full crawl after installation, we are using Adobe ifilter – SPAdmin Jun 9 '15 at 10:02
0

If you have a lot of scanned, faxed or other kind of image based content then you will need to carry out a separate OCR step for best results. This will place a hidden layer inside the PDF with all the text inside.

Any modern search crawler, including SharePoint's, can then index the content of the documents resulting in many documents, that were previously un-indexable, to show up in search results.

Added bonus, you can copy and paste text, as well as search inside image based PDFs using any PDF reader.

Third party software is available to automate the OCR process using a simple workflow.

Full disclosure, I worked on such as project. For details see this Knowledge Base Article.

Improve this answer
0

It is entirely based on the OCR software that created the PDF (and added the discovered text). In SharePoint 2013 PDFs are indexed natively. In SharePoint 2010 you have to add a third party. You indicate that you are using Adobe, personally I would NEVER use the Adobe PDF iFilter. You will get significantly better results from foxitsoftware.com and their PDF iFilter. It is significantly FASTER than Adobe, so your crawls will finish faster. It is SUPPORTED, and yes, you have to pay for it.

Improve this answer
  • I agree with you. I would probably consider foxit. – SPAdmin Jun 3 '15 at 21:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.