View unanswered posts | View active topics


Reply to topic  [ 12 posts ] 
Author Message
  
 Post subject: wide area motion detection and tracking
PostPosted: Sat Jan 03, 2009 12:09 pm 

Joined: Jan 2009
Posts: 10

Offline
What is the best wide area motion detection and tracking video surveillance system?Please advice us best and modern system in real time motion tracking and excelent manufactore in the world?


Top
 Profile  
 

  
 Post subject:
PostPosted: Sun Jan 04, 2009 1:10 pm 

Joined: May 2004
Posts: 117

Offline
virtually all of the systems today use the same technology which is pixel changes. So all will be the same in this regard. The only time this changes is when you get into the intelligent video which determines which is real motion and what it is. In other words, you don't want blowing trees or debris setting off motion alarms. When you get into the intelligent video analytics it gets very expensive.


Top
 Profile E-mail  
 

  
 Post subject:
PostPosted: Mon Jan 05, 2009 11:48 am 

Joined: Jan 2009
Posts: 10

Offline
Thank you for reply
could you tell us famous company in intelligent video system,especially in wide area real time motion detection and tracking?


Top
 Profile  
 

  
 Post subject:
PostPosted: Thu Jan 08, 2009 1:29 pm 

Joined: Jan 2009
Posts: 12

Offline
I highly recommend the VideoIQ products as you can set parameters for what you want the camera to look for. It turns on only when the progamed event happens saving bandwidth and need for someone to watch over the screens 24/7. It will then send out an alert of a breach of that parameter. PM me for more info.


Top
 Profile  
 

  
 Post subject:
PostPosted: Sun Jan 11, 2009 7:33 am 

Joined: May 2004
Posts: 117

Offline
hardware based systems such as the videoIQ are a problem. Since the core parameters are "fixed" you have little ability to teach the system to learn patterns and be able to exclude known items that would create false alarms.


Top
 Profile E-mail  
 

  
 Post subject:
PostPosted: Sun Jan 11, 2009 7:36 am 

Joined: May 2004
Posts: 117

Offline
simple examples of hardware based systems with fixed parameters would be:

Lets say you are interested in left objects around an airport for obvious reasons. There are luggage trolleys everywhere on an airport. People strand them everywhere. If everytime one was stranded an alarm went off then the system would become useless. So you would need a method first to identify what a luggage trolley is and then the difference between a full trolley and an empty trolley. This can only be done with software systems on the fly as opposed to hardware systems which would require major headaches.


Top
 Profile E-mail  
 

  
 Post subject:
PostPosted: Tue Jan 13, 2009 1:43 pm 

Joined: Jan 2009
Posts: 12

Offline
cctvexpert wrote:
hardware based systems such as the videoIQ are a problem. Since the core parameters are "fixed" you have little ability to teach the system to learn patterns and be able to exclude known items that would create false alarms.


Here is a real life install with the Video IQ cameras and how they are limited in their "False Alarms".

Highland Builders in the Boston, MA area has deployed 1-4 iCVR cameras at their higher risk sites to monitor daytime activity and generate alerts for activity on the property after hours. Many of their clients have large properties in areas where there is wildlife activity. This, along with some construction debris that can blow in the wind, prevents the use of simple video motion detection, as the false alarm rate would be excessive.

As construction progresses, the camera mounting location is sometimes moved to a different point on the property. A solution that required constant recalibration would be cumbersome to maintain. The iCVR camera installs quickly, requires no complex setup, and provides a reliable mechanism to receive after-hours alerts or to allow the builder to check on his crews throughout the day.


Top
 Profile  
 

  
 Post subject:
PostPosted: Tue Jan 13, 2009 10:43 pm 

Joined: May 2004
Posts: 117

Offline
how do you differentiate between animals and humans
how do you account for changes in weather
how do you account for lighting issues
how do you calibrate (no analytics is plug and play) its an algorithm


Top
 Profile E-mail  
 

  
 Post subject:
PostPosted: Thu Jan 15, 2009 10:03 am 

Joined: Jan 2009
Posts: 12

Offline
This is how our rep from VideoIQ replied. If you want to learn more, please contact me and I would be glad to get you in touch with the folks over at VideoIQ.


1) Humans vs Animals - That’s what we do. The product is extremely good at properly categorizing non-human objects. Occasionally a large animal (deer, etc) may be temporarily mis-identified as a human if it is approaching the camera head-on, this is usually not an issue when combined with other rule or ROI factors (such as a dwell time in combination with an ROI). VideoIQ’s approach is different. We look for patterns in the pixels that match the objects we are trying to detect.

*Remember that our current analytic implementation is for – people, vehicles, suspicious objects. This is our core competency; we have made it plug and play.

2) Weather and 3) Lighting - Our analytics algorithms have been specifically adapted over 7 years of development to be immune to the factors that affect other video analytics solutions. This includes lighting changes, weather (snow, rain, etc. in the field of view causing lots of pixel changes), and gradual scene changes (ie: seasonal changes of leaves on trees, growth of foliage, etc). Our analytics self-calibrate by observing moving objects in the scene and comparing them to our pattern database.

Because we use a pattern-match approach, things like lighting weather, etc are not a factor, as these issues do not alter the overall “patternisticâ€

_________________
WAV, Inc
Specializing in Wireless and IP Cameras
Visit us at www.wavonline.com


Top
 Profile  
 

  
 Post subject:
PostPosted: Fri Jan 16, 2009 9:36 am 

Joined: Jun 2008
Posts: 35
Location: Atlanta, Ga

Offline
I have been playing with the system from Aimetis. It's been stable and I'm still learning but it seems like a well developed product. Pricing is by camera in Standard, Professional, and enterprise versions. Can get kind of expensive. www.aimetis.com :)


Top
 Profile E-mail  
 

  
 Post subject:
PostPosted: Mon Jan 19, 2009 7:41 am 

Joined: May 2004
Posts: 117

Offline
There is no such thing as "self learning" unless you have invented a neural intelligent algorithm which i doubt is the case since the processing power required would take up 100 cameras.

People should be honest about analytics and tell it like it is. I have been working with analytics for a decade now and there are simply things it will do and wont do and hardware analytics do suffer in a number of areas.

As far as having it self calibrating and self learning we can get into a whole discussion of math and science why that is not possible.

And to even use the words plug and play with video analytics well shame on you. We have tested virtually every product out there over the years and watched a number of shoot outs by public agencies and they all fail in the same areas.

That being said, analytics is a great tool but not a panacea. It will never be plug and play as each and every scene is different, each and every shadow is different, lighting is different, background activity is different, etc.

Even the best software algorithms which do true scene content analysis and are light years ahead of the objectvideos of the world still need extensive calibration and optimization.

If it is so good can you give 3 references of people using it successfully in an enterprise environment. There is your answer. Talk to the customers using it system wide.


Top
 Profile E-mail  
 

  
 Post subject:
PostPosted: Wed Jan 21, 2009 3:41 pm 

Joined: Jan 2009
Posts: 12

Offline
I have 5 more real world applications I can give you, but I don't have the time to type them all out. They are for a Car Ferry Boat, Cruise ship and Cargo, commericial buildings and property owners, chemical/gas facility, and residential communites.

_________________
WAV, Inc
Specializing in Wireless and IP Cameras
Visit us at www.wavonline.com


Top
 Profile  
 

Display posts from previous:  Sort by  
Reply to topic  [ 12 posts ] 


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  

It is currently Thu Sep 02, 2010 2:24 pm
Custom Skin by Audentio Design.

The contents of this webpage are copyright © 2003-2010 CCTVForum.com. All Rights Reserved.