How reliable is the VideoIQ analytics

[serverguy 0]

Hi all, been a while since I have been on here, see things have moved on a bit and IP cameras falling in price

 

I have a job to install some cameras at somewhat remote farms. Generally large areas of concrete surrounding a shed with some machinery parked about the place.

 

The original idea was just to install cctv that records 24/7 but I am reluctant to go down that route a) because it required lots of cameras to cover an outdoor area effectively, b) because 99% of the theft would occur at night and at night getting an image that would be useful in investigation is even more difficult and c) the logistics of streaming lots of IP cameras back to a central DVR over a wireless connection is not easy.

 

I am very reluctant to install a cctv system that just shows some ghost stealing scrap and driving off in it. Numberplate capture is useless because the plate is stolen. CCTV like that is completely useless and a total waste of money - and this is what 95% of cctv systems provide.

 

So I have come across the VideoIQ stuff. Considering that the farms will be very (read VERY) low traffic areas, and the only humans likely to visit will be people who should not be there, how many false alarms am I going to get?

 

My idea is a couple VideoIQ megapixel cameras giving an overview of the important areas of the farm and if it recognises a human fire off an email (or we are investigating remote monitoring).

 

We have toyed with PIR sensors and trip wires etc but they all add hassle and complexity. The idea of distributed storage with VideoIQ is very nice as they will work well with the mesh networks we install.

 

Secondly, by the time you buy decent IP camera + decent recording server + network / cable infrastructure + PIR / beam sensors the VideoIQ seem very attractive especially if they can work proactively.

 

I am going to puchase one to have a play with it but basically, how reliable / accurate is the detection (in favorable circumstances - basically zero traffic / floodlit) or would we just be plagued with false alarms every night?

 

Thanks for any insight.

[tomcctv 190]

I am very reluctant to install a cctv system that just shows some ghost stealing scrap and driving off in it. Numberplate capture is useless because the plate is stolen. CCTV like that is completely useless and a total waste of money - and this is what 95% of cctv systems provide.

 

 

 

it just depends on how its installed and the quality of the equipment used.

 

So I have come across the VideoIQ stuff. Considering that the farms will be very (read VERY) low traffic areas, and the only humans likely to visit will be people who should not be there, how many false alarms am I going to get?

 

 

 

you can get DVRs/NVRs that will do just the same and on a farm you will get some bad alarms.

 

the logistics of streaming lots of IP cameras back to a central DVR over a wireless connection is not easy.

 

if you are looking at IP then no need for a DVR but you are looking at some good money to build a good wireless network .

 

 

how many cameras are you looking at what is the size of your area to cover and the big one. what is your budget.

[thewireguys 3]

I have deployed VideoIQ cameras and for what you want to do it will be a good fit. I have found that VideoIQ for preventing/deterring crime before it happens IF you have someone that is live monitoring the system. You can also setup 2-way auto and remote trigger I/O for triggering bells and sirens. Their analytics are self calibrating so you can deploy them fast and without hours of labor to configure the analytics.

[serverguy 0]

Hi tom

 

Will try to answer your questions

 

I am very reluctant to install a cctv system that just shows some ghost stealing scrap and driving off in it. Numberplate capture is useless because the plate is stolen. CCTV like that is completely useless and a total waste of money - and this is what 95% of cctv systems provide.

 

 

 

it just depends on how its installed and the quality of the equipment used.

 

The figures aren't good for cctv Only 1 in 1000 crimes in UK that get recorded on a cctv camera get solved

 

And the fact that the police know full well which group of people who do all the thefts are anyway (gypsies...) and still fail to do anything about it whatsoever I would say for rural security just recording for playback after the crime is a bit of a chocolate fireguard. The only time the police are interested is if they can catch them in the act.

 

So I have come across the VideoIQ stuff. Considering that the farms will be very (read VERY) low traffic areas, and the only humans likely to visit will be people who should not be there, how many false alarms am I going to get?

 

 

 

you can get DVRs/NVRs that will do just the same and on a farm you will get some bad alarms.

 

I was under the impression the VideoIQ stuff was the best of the bunch when it came to analytics especially at a price point - I could be wrong however?

 

I also like the idea of the VideoIQ cameras because of the 'edge' recording - having all the recording done on a central NVR would mean lots of traffic on the network all the time. I much prefer the idea of edge recording and the network only being used when it is sending an alarm or archiving.

 

the logistics of streaming lots of IP cameras back to a central DVR over a wireless connection is not easy.

 

if you are looking at IP then no need for a DVR but you are looking at some good money to build a good wireless network .

 

Surely if the camera does not have onboard storage there is going to have to be a storage box of some sort somewhere to record.

 

We install wifi networks in awkward places / build temporary networks at public events / shows. The cost of wifi kit is peanuts compared to IP cameras! Of course it depends what you use, if you are using something like Ruckus or Fluidata its a lot more.

 

 

how many cameras are you looking at what is the size of your area to cover and the big one. what is your budget.

 

 

Not massive areas. A typical example would be enormous shed (100 x 50 metres) with a 20 x 50 metre office / workshop opposite it with a 30m concrete area between them where 99% of the action is and the access to workshop / office etc.

 

My thoughts would be to have a megapixel camera looking at this area and if it sniffs out a human send an email.

 

We have experimented with other ways of doing it and have had quite good success with infared beams to act as the alarm triggers and then a camera to have a look.

 

However, by the time you price

 

500 quid for a camera

250 quid for a box to record the camera

300 quid on infared beams

100 quid on networking kit

100 quid on a UPS

Lots of £££ time setting it all up

 

And take into account the inevitable problems of tractor parked in front of the beams / someone knocks the beams off with tractor bucket / recording software crashes etc etc

 

The idea of 1300 quid on a camera that 'just works' is quite appealing

 

 

Thanks

Edited February 21, 2012 by Guest

[serverguy 0]

I have deployed VideoIQ cameras and for what you want to do it will be a good fit. I have found that VideoIQ for preventing/deterring crime before it happens IF you have someone that is live monitoring the system. You can also setup 2-way auto and remote trigger I/O for triggering bells and sirens. Their analytics are self calibrating so you can deploy them fast and without hours of labor to configure the analytics.

 

Thanks

 

That is what I am thinking - camera sees person - sends alarm for farmer - farmer looks at ipad and sees pikey - farmer leans out window and shoots pikey (of course the last part is said tongue in cheek!)

 

Or farmer hears alarm - looks at ipad - sees ghost inspecting the camera - farmer goes back to sleep (hopefully there won't be too many of these)

 

We already put a speaker out with the camera and that works really well.

 

I am looking at some sort of live monitoring but the cost most of the monitoring centres want is prohibitive for an alarm that may only go off a handful of times a year.

[NotoriousBRK 0]

Hi...

 

I run the Field Engineering group for VideoIQ worldwide. My job is basically to make sure things work in real life and that the sales people stay honest

 

First off, you're not going to get anywhere near the quality/reliability of this solution with PIR's, DVR/NVR software, "advanced motion detection" or any of the other relatively unsophisticated "solutions" on the market.

 

We're also the *only* ones doing real megapixel analytics, meaning truly analyzing the full stream, not downsampling it and running an analysis on sub-sampled images.

 

Your application fits the description of probably 80% of our typical deployments. Not the bit about the farm, but the bit about having a tricky location where you need something that can truly help deter crime.

 

I've got a really good distributor based in the UK (where it seems like you are). PM me your info and I'll put you in touch with them. I'll also give you my corporate contact info and you're welcome to contact me as well with any questions or whatever.