Create searchable (Images) in PDFs via the SDK (by using .Net)

[Jörg_Battermann@adobeforums.com]

Good afternoon,

we're currently starting to roll-out a (pdf) document generation solution based on XSL-FO transformation from xml to pdf files which all works perfectly fine, but one current constraint is that embedded images are not searchable by the users.

Now I understand that you can in fact do that manually when having Acrobat Professional installed and using it's Text Recognition/OCR functionality, but I was wondering whether the same technology is available through the SDK.

As I haven't worked with the SDK myself, yet, I'd also like to know how well it performs when being used in .Net/C#-based applications, are there any specific constraints, limitations that this environment would cause?

The way I would want to implement it is to add an (optional) post-processing step after the xsl-fo transformation (which takes place on a central server with a queue etc) which would perform this text recognition, but since we create a couple hundred thousand pages each day, reliability and scalability are a big factor. One document-generation machine/server currently creates a couple documents in parallel on a standard multicore machine with a couple gigs of ram, but has anyone tested the sdk's functionality under heavy load extensively?

I am just double-checking because we've hit several walls in the past with such under-communicated limitations of 3rd party vendors....

Cheers and thanks,
-Jörg Battermann

[PDL@adobeforums.com]

but I was wondering whether the same technology is available through the
SDK.




Yes, because the SDK only automates Acrobat. It is not a self-contained SDK or redistributable - applications developed with the SDK require a copy of Acrobat installed on the same machine in order to run.

The way I would want to implement it is to add an (optional) post-processing
step after the xsl-fo transformation (which takes place on a central server
with a queue etc) which would perform this text recognition




Then you will want to look at a different product, since Acrobat's EULA prevents it from being installed on a server as part of a purely automated workflow.

but since we create a couple hundred thousand pages each day, reliability
and scalability are a big factor.




Again, Acrobat is not technically suitable for this type of environment.

but has anyone tested the sdk's functionality under heavy load extensively?




No, since again the SDK only automates Acrobat, and Acrobat is not technically suited nor licensed for this type of high-availability server environment.

Have you contacted Adobe to find out if any of their LiveCycle Server products will suit your needs? LiveCycle PDF Generator has OCR functionality (not sure if it will directly suit your needs - you should contact Adobe directly for that answer), but I know it can OCR filetypes like TIFF, and it was specifically designed to work in a high-availability server environment.