Re: Video Surveillance Requirements

[randy marchany <marchany () VT EDU>]

On Thu, Jan 14, 2010 at 1:57 PM, Patrick Ouellette
<ouellep () algonquincollege com> wrote:
> I'd go one step further - since we don't have explicit sign-off to display people in the video, there is a privacy 
> issue.
> However, there is software that will allow for auto-blurring faces and key elements in the video as it's fed back out 
> from the storage, but keep the original input feed entirely clear.

There are a number of issues that you need to consider before going
through with this type of project. Some of the following is from my
experience of dealing with video surveillance vendors after the APril
2007 incident here.

1. FERPA - relates to the privacy comment above. I'm sure you have
some students who have a "privacy" tag on their records for whatever
reason. Not having a sign informing everyone that they are under
surveillance may cause problems if their desire to remain
private/anonymous is for safety reasons (stalker, parent with a
restraining order against them, etc.). I believe it's prudent to
inform the occupants that they may be under surveillance. Some people
don't care, some do. I'm not saying you can't film public areas but
from a PR standpoint, informing them addresses a lot of issues.

2. Any video camera system must restrict who can access its images.
See webcam.prejeans/com/view/index/shtml and ask yourself if you
really want the entire internet to see these images.

3. Any video camera system must encrypt the video feed from the
cameras to whatever server. Again, should there be an incident, you
don't want a video of the crime to be on YouTube or CNN. Having the
video cams on a physically separate network isn't secure. This is an
issue if you're looking at wireless camera systems. Wireless is radio
and signals can be intercepted and replayed.

Just my .02.

Randy Marchany
VA Tech IT Security Office