Data Loss Prevention

Hi all,

Is there an option to block Image file ( screenshot for example ) that contain ip address or credit card number ?

Thanks,

Hi Benda,

Doesn't seem like DLP has that functionality. Check below:

https://www.symantec.com/connect/forums/how-protect-sensitive-information-image-file

It would have to be a block of all images, not the information contained therein.

Thanks!

Hi Benda

OCR may be what you're looking for. OCR can learn to read images and pick out characters in line with your policies. There is also form recognition that can do the same for documents in PDF etc. If you were to use OCR with a policy, a response rule can be to quarantine the image?

See this article as OCR was introduced in 15.0 https://support.symantec.com/en_US/article.TECH221405.html

Hope I've understood you correctly.

Please mark as resolved if appropriate.

Thanks

Chris

Hi Is the ocr an add on?  From the KB symantec OCR appears to have limits, anyone know it's limitations?  

Thank you,

Jyl

Hi Jyl

I believe OCR is configured as part of the DLP product as part of releases from 15.0 onwards. I've configured it and not had to add anything on to the product.

There are certain conditions (like making sure the PDF is a certian quality etc) but you would have to be more specific about this.

Thanks

Chris

thank you, we'll test it

Great thanks

Please be sure to mark my comment as the solution if this is what you needed.

Thanks

Chris

Hi there,

As I understand it - Sensitive Image Recognition (OCR) has an additional license requirement, however depending upon your subscription - it may be included in your suite of products. Please chat to your Account Rep to confirm

With versions 15.0, 15.1 and 15.1 MP1 - one limitation worth calling out is that images will not be detected in Microsoft Office products. This has now been resolved with version 15.5

OCR is configured by attaching an OCR Detection Engine (a separate OCR server running on Windows Server) to a Network Prevent or Discover server.

General sizing requirements are 1.25 OCR servers per Detection server (however this is dependant upon traffic throughput)