Path: news.mathworks.com!newsfeed-00.mathworks.com!newsfeed2.dallas1.level3.net!news.level3.com!postnews.google.com!p25g2000hsf.googlegroups.com!not-for-mail
From: ImageAnalyst <imageanalyst@mailinator.com>
Newsgroups: comp.soft-sys.matlab
Subject: Re: human tracking
Date: Wed, 2 Apr 2008 12:19:05 -0700 (PDT)
Organization: http://groups.google.com
Lines: 80
Message-ID: <e4812e26-e7dd-465c-ba58-848e9d6f95f1@p25g2000hsf.googlegroups.com>
References: <fstnms$965$1@fred.mathworks.com> <9488d7ca-866e-49e7-8307-88d3845706d6@8g2000hsu.googlegroups.com> 
NNTP-Posting-Host: 192.44.136.113
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
X-Trace: posting.google.com 1207163946 28765 127.0.0.1 (2 Apr 2008 19:19:06 GMT)
X-Complaints-To: groups-abuse@google.com
NNTP-Posting-Date: Wed, 2 Apr 2008 19:19:06 +0000 (UTC)
Complaints-To: groups-abuse@google.com
Injection-Info: p25g2000hsf.googlegroups.com; posting-host=192.44.136.113; 
User-Agent: G2/1.0
X-HTTP-UserAgent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1; .NET 
X-HTTP-Via: 1.1 bdci2px (NetCache NetApp/6.0.6)
Xref: news.mathworks.com comp.soft-sys.matlab:460667



On Apr 2, 2:29=A0pm, "Chee Sing Wai" <williams_lib2...@hotmail.com>
wrote:
> > Could be very difficult. =A0Could be a Ph.D. project instead
> of just a
> > simple final year project. =A0So you have objects moving in
> your video
> > and you need to detect whether they are human versus car,
> truck, dog,
> > bird, etc. =A0You better be able to make a lot of
> assumptions or it's
> > going to be very difficult. =A0For example, let's say you
> have a small
> > blob moving across the top of your frame. =A0How do you know
> it's a bird
> > (or a plane) and not a human that is far away? =A0You can't
> look for
> > skin color because the person could be walking away and
> not showing
> > any skin. =A0Maybe you can assume that anything moving in a
> certain zone
> > is human (say along a sidewalk) and anything moving in
> another zone
> > (say, the street) is a vehicle. =A0You might say that
> vehicles move
> > faster than humans but of course we know that this is not
> always the
> > case. =A0Maybe you could do some sort of shape recognition
> but this is
> > going to be very hard. =A0Maybe you could pick out people
> based on the
> > size of the blob, but this assumes everything is about the
> same
> > distance from the camera. =A0For a simple example, if you
> were looking
> > at a crosswalk with a limited field of view that you had
> to monitor,
> > then you could assume that if its size is in a certain
> range, and its
> > speed is in a certain range, then it might be human.
> Maybe you could
> > post a few frames (on a web site somewhere), list your known
> > parameters, restrictions, and assumptions, and then ask
> the smart
> > folks over in sci.image.processing for tips.
> > Regards,
> > ImageAnalyst
>
> @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
>
> Thanks ImageAnalyst for the reply.
>
> Then lets say that my scope is only limited to systems such
> as indoor surveillance system (e.g. a surveillance camera
> watching a room or corridor..) and I am only tracking a
> single moving object. Then how well will the outcome be if I
> determine whether the moving object is human or not base on
> the size of the blob?
>
> Haha.. actually I have been trying to do this using some
> sort of shape or pattern recognition but just like you have
> mentioned, it is quite difficult to do it due to lacked of
> knowledge and examples in this field. It would be great if
> anyone can provides some useful links or examples regarding
> the related field. Thanks- Hide quoted text -
>
> - Show quoted text -
----------------------------------------------------------------------------=
--------------------------------
Chee Sing Wai:
What other single object do you expect to see moving through the
room?  A dog?  A robot?  A paper airplane?  I would think in a very
simplified and controlled final year project where you can control
what goes through the room, then doing it based on size should do very
well.

Here's a useful link for you:
http://www.google.com/search?hl=3Den&q=3Dhuman+tracking+surveillance
Looks like there's lots of good stuff there.
Good luck,
ImageAnalyst