Bank/Financial System

[CoolDude 0]

Looking for suggestions on the specs for cameras and DVR to capture the best images for a bank. Most cameras will have a fairly broad field of view with at least two cameras at a choke point to capture faces entering and exiting. Prefer stand alone with LAN network capability. Wireless or internet access is not an option for security reasons. Minimum 16 cameras with second DVR to 32.

[VST_Man 1]

High End

 

http://www.exacq.com/

 

with

 

http://pro.sony.com/bbsc/ssr/product-SNCDM160/

 

and

 

http://www.arecontvision.com/

 

hard wire all, use a seperate network, enjoy the pics..

[Erron S. 0]

Cooldude, I think I may be able to help a little on this. I'd bet they have the restrictions on their network compliance which is why they can't have the dvr put on their network. I bet they have to have a PCI (Payment Card Industy) complient network. Not only banks but many retailers are going through these changes as well. Take a read here..

http://www.3xlogic.com/pci

 

In addition to the Exacq, have a look see at our as well, especially if you are doing megapixel with extended record rates. As far as cameras, I've been leaning more and more toward the IQ Eye line of cameras. www.iqeye.com

They have a few features that make them stand out from some of the others I've seen. The IQ focus feature also you fine tune the lens so you can get every bit of sharpness from the camera that's possible. As we all know, getting an MP camera lens perfectly focused is a little tricky without a hand held device and features like this makes installation a little easier. Hope that helps a little.

[Soundy 1]

Cooldude, I think I may be able to help a little on this. I'd bet they have the restrictions on their network compliance which is why they can't have the dvr put on their network. I bet they have to have a PCI (Payment Card Industy) complient network. Not only banks but many retailers are going through these changes as well. Take a read here..

http://www.3xlogic.com/pci

 

In addition to the Exacq, have a look see at our as well, especially if you are doing megapixel with extended record rates. As far as cameras, I've been leaning more and more toward the IQ Eye line of cameras. www.iqeye.com

They have a few features that make them stand out from some of the others I've seen. The IQ focus feature also you fine tune the lens so you can get every bit of sharpness from the camera that's possible. As we all know, getting an MP camera lens perfectly focused is a little tricky without a hand held device and features like this makes installation a little easier. Hope that helps a little.

 

I'll second all that... 3xlogic makes some really nice DVR/NVR systems that I've used in some high-demand applications. Biggest one so far is a restaurant with 23 analog cameras and five 1.3MP IQ-511 cameras, along with a grand total of almost 10TB of storage.

[jhonovich 0]

Almost no bank uses megapixel cameras for 2 reasons:

 

- The cost of the cameras are too high

- The cost of the storage is too high

- Bandwidth problems for remote viewing (most banks have 256 Kb/s or less connections - even low frame rate megapixel cameras will create significant problems here)

 

I am not sure what type of budget the bank you are dealing with has. For most banks, the budgets are tight and won't accommodate megapixel cameras.

 

March and Verint (Lanex NetBox) are the two biggest banking DVR providers (by sales).

 

I don't know enough about your particular situation to make a strong recommendations but I would encourage you to consider these factors.

[VST_Man 1]

times are changing........and so is crime.................megapixel camera prices are droping and stroage is cheap, which allows "us" to compete with better pics at a "slight" increase in cost.

 

Don't discount the option when selling................sell both systems and allow the Bank to decide. You'll look better.

 

Besides, I heard the Banks just some spending money.........

[Erron S. 0]

Almost no bank uses megapixel cameras for 2 reasons:

 

- The cost of the cameras are too high

- The cost of the storage is too high

- Bandwidth problems for remote viewing (most banks have 256 Kb/s or less connections - even low frame rate megapixel cameras will create significant problems here)

 

I am not sure what type of budget the bank you are dealing with has. For most banks, the budgets are tight and won't accommodate megapixel cameras.

 

March and Verint (Lanex NetBox) are the two biggest banking DVR providers (by sales).

 

I don't know enough about your particular situation to make a strong recommendations but I would encourage you to consider these factors.

 

The world is a' changing though. The banks I've quoted, speced, and sold have begun to go the way of the newer technologies because the conviction rate and image quality has been so poor in the past.

 

Cost of megapixel? http://www.3xlogic.com/blog

Cost of Storage? Tiger Direct has their 1.5 TERABYTE drives for sale at $119. That's if you buy one, I wonder what the bulk rate is direct from Seagate?

You have to recompress megapixel to transmit it effectively or you're exactly right, the bandwidth just isn't there yet. Check out the graph at the bottom of this page.

http://www.3xlogic.com/megapixel

[jhonovich 0]

Hi Erron,

 

Thanks for the feedback. I have discussed your Aztech codec with Dave Nieweg a few months back and as I told him I think it's valuable. If you want to create a post just on Aztech for my site, feel free.

 

That being said, Aztech and Avigilon are exceptions in megapixel bandwidth management. Almost all NVR providers have problems streaming megapixel cameras over low bandwidth links.

 

As for your cost of megapixel analysis:

 

- NVRs are cheaper than DVRs. Please be specific on how much cheaper. I would estimate $500 to $1000 cheaper from eliminating the capture card. However, megapixel cameras are ~ $500 more each. The total cost is still far more expensive.

- Cat 5 is cheaper than coax. In a bank with such short cable runs, the difference in cost is negligible (plus most banks already have coax in place).

- Megapixel's higher resolution: You can reduce camera count but then you are not really increasing the image quality. This is fine but you can't have both.

- While hard drive costs continue to fall, what is the total amount of storage that megapixel camera bank system needs? what does that cost (not the hard drive, the end user cost for the entire storage system)? Banks with 16 channels can easily get away with 500GB for analog camera using CIF. With megapixel cameras, you could easily require 4GB or more.

 

I have worked on deals with dozens of banks and almost all of them have tight budgets for video surveillance systems. A megapixel based system is certainly going to cost a thousand if not thousands more. I am skeptical how many will pay the premium, especially in the light of today's economy.

[Soundy 1]

As for your cost of megapixel analysis:

 

- NVRs are cheaper than DVRs. Please be specific on how much cheaper. I would estimate $500 to $1000 cheaper from eliminating the capture card. However, megapixel cameras are ~ $500 more each. The total cost is still far more expensive.

 

A *quality* analog camera will easily run $250-$500 on its own. Cameras do fade and fail over time, so if you're talking about replacing an old system, factor in that it may be time to replace their cameras anyway. If you can get four times the resolution at less than twice the price... that's a pretty easy sell.

 

- Megapixel's higher resolution: You can reduce camera count but then you are not really increasing the image quality. This is fine but you can't have both.

 

It's still better than a 2:1 ratio - one IP cam can often replace *three* analog cameras. Speaking specifically from fuel services experience, I can cover 3-4 pump lanes (both sides of two islands) and read license plates easily with a single 1.3MP IP camera... the same thing with analog cameras would take up to four cameras, each with fairly tight focus, so there's less ancillary coverage. The IP cam is an easy sell in this case.

 

- While hard drive costs continue to fall, what is the total amount of storage that megapixel camera bank system needs? what does that cost (not the hard drive, the end user cost for the entire storage system)? Banks with 16 channels can easily get away with 500GB for analog camera using CIF. With megapixel cameras, you could easily require 4GB or more.

 

Once again... more data per camera, but fewer cameras required. It's as much a non-factor as cabling costs.

 

I have worked on deals with dozens of banks and almost all of them have tight budgets for video surveillance systems. A megapixel based system is certainly going to cost a thousand if not thousands more. I am skeptical how many will pay the premium, especially in the light of today's economy.

 

I've found a really easy way to sell IP cameras: show them the picture. One pub/liquor store customer we have, I was in talking to the owner, was telling him about another nearby job we were doing, and showed him a set of stills from the IP cams we'd just put in there.

 

His immediate response: "I want that!" He ordered two for that store and one for another of his stores, and I subsequently added another to the first store.

 

Once customers see these things, there's just no going back for them.

[jhonovich 0]

From a security design perspective it is not possible to replace 4 analog cameras with a single 1.3 MP camera.

 

Each analog camera is 640 x 480. A 1.3 MP camera is 1280 x 1024. While that is 4x the number of pixels, it is only 2x the width of FOV.

 

For almost all surveillance applications, including lanes at a gas station or teller positions at a bank, it is the number of pixels wide that is the key consideration. Almost all applications have a very narrow range of height to cover. As such, a 1.3 MP camera can only cover twice the target area as an analog camera. If a 1.3 MP camera replaces more than 2 analog cameras, you are essentially guaranteed to have lower resolution than an analog camera.

[Soundy 1]

Sorry, but my experience disagrees with your calculations.

[jhonovich 0]

Sorry, but my experience disagrees with your calculations.

 

This sums up my general problem my most megapixel advocates. Their advocacy often boils down to:

 

1. Look at the pretty pictures.

2. Cost is no object.

 

I apologize for venting my frustration. However, each time I offer calculations and detailed cost estimates, commenters refuse to engage on a detailed and substantive level.

 

I think megapixel cameras are the most exciting and important trend in the video surveillance industry. However, megapixel cameras have real issues - that's why less then 3% of all camera sales this year are megapixel - and it's not simply because people are too stupid to use them. I am happy to discuss these topics but please respond to my detailed points and calculations. If you think my calculations on cost or performance are incorrect, I am happy to be corrected on the facts.

 

Best,

 

john

[Soundy 1]

What sort of detail would you like me to engage on?

 

I put cameras on fuel service sites. One 1.3MP camera, depending on mounting location, can easily replace three analog cameras for monitoring pumps. This is my experience. On a cost level, that's three *good-quality* analog cameras, along with lenses and the requisite wiring. Camera, lens and wiring together can easily run upward of $500+, each. That's $1500 worth of cameras, replaced by a single camera that, with lens and installation, runs barely half that.

 

Your calculations are not inaccurate... but they're not the whole story either.

[cool guy 0]

What sort of detail would you like me to engage on?

 

I put cameras on fuel service sites. One 1.3MP camera, depending on mounting location, can easily replace three analog cameras for monitoring pumps. This is my experience. On a cost level, that's three *good-quality* analog cameras, along with lenses and the requisite wiring. Camera, lens and wiring together can easily run upward of $500+, each. That's $1500 worth of cameras, replaced by a single camera that, with lens and installation, runs barely half that.

 

Your calculations are not inaccurate... but they're not the whole story either.

 

I am Research Engineer working in Video Analytics area and I often deal with these kind of situations. Isn't it good enough to use the existing camera network rather than upgrading to Mega-pixel cameras for using it for Video Analytics etc? If you are just trying to record video continuously without any video analytics, well I don't know what to say (old school, maybe??)

 

But given your experience in this area, when you are investing so much money on video surveillance, isn't it a smart idea to think ahead and implement a system that can scale well for future applications. In which case Megapixel cameras is your ANSWER!!!

 

When designing a video analytics algorithm (say License Plate Recognition), there are certain pre-requisites like the pixel height of the License Plate text in the image, quality of the image etc which have to be taken into consideration for accurate recognition. With 1.3MP analog cameras, it might work best if you can focus the cameras for individual lanes. Just because a human can guess the text license plate from a 1.3MP camera covering 3 lanes (which I highly doubt), doesn't mean that the computer can recognize it too. Humans have a higher level of visual perception than computer algorithms. It is a Holy Grail for any algorithm designer to replicate human vision on a computer and is still many years away.

 

If you can upload a sample image from the 1.3 MP analog camera covering 3 lanes of traffic and still read the text on License Plate, your arguments can be taken seriously. But I honestly don't believe that's possible.

[Erron S. 0]

I do agree with the 3% of all camera sales are megapixel. Consider this though. That includes all cameras, from the Sam's Club specials to the $35 IR camera from Harbor Freight. While these cameras can and will tell you when you're buddy pulls up in front of the house, it's not going to give you a plate number or a clear enough shot to give to the news station and police to put out an APB. There are a number of reasons to use MP over analog (which I kind of touch on it my blog) but really the main reason I see business owners buy and use MP cameras is clearly for the image quality. If they are going to spend X thousands on a full blown system they really need to be able to identify what they are looking at and have useable video. This could be a criminal, money denominations, a license plate, or the difference between someone putting something into their pocket or scratching an itch. I would also say that a number of MP cameras are going into locations that have had analog for years and they have had their frustrations with the image quality along the way. It's typically the difference between 'the guy with the blue hat' and the 'the guy with the blue hat and UCLA logo on it'. The absolute best way to sell it is just show it. I didn't want an HD TV until I -saw- one. I never thought standard TV to be all that bad, but now normal TV is hard to watch in comparison. Know what I mean?

[Soundy 1]

I am Research Engineer working in Video Analytics area and I often deal with these kind of situations. Isn't it good enough to use the existing camera network rather than upgrading to Mega-pixel cameras for using it for Video Analytics etc? If you are just trying to record video continuously without any video analytics, well I don't know what to say (old school, maybe??)

 

Well, the only "analytics" being dealt with in most of these cases are carbon-based optical-brain interface... ie. people looking for detail in video playback

 

But given your experience in this area, when you are investing so much money on video surveillance, isn't it a smart idea to think ahead and implement a system that can scale well for future applications. In which case Megapixel cameras is your ANSWER!!!

 

Absolutely. In fact, I try to wire Cat-5e whenever possible on new installs, rather than using coax, specifically for "futureproofing". If they just want analog cameras for the time being, a pair of video baluns will do the job nicely, and they're ready to upgrade to IP at any time!

 

When designing a video analytics algorithm (say License Plate Recognition), there are certain pre-requisites like the pixel height of the License Plate text in the image, quality of the image etc which have to be taken into consideration for accurate recognition. With 1.3MP analog cameras, it might work best if you can focus the cameras for individual lanes. Just because a human can guess the text license plate from a 1.3MP camera covering 3 lanes (which I highly doubt), doesn't mean that the computer can recognize it too. Humans have a higher level of visual perception than computer algorithms. It is a Holy Grail for any algorithm designer to replicate human vision on a computer and is still many years away.

 

Sure, but where do you find 1.3MP analog cameras??

 

 

If you can upload a sample image from the 1.3 MP analog camera covering 3 lanes of traffic and still read the text on License Plate, your arguments can be taken seriously. But I honestly don't believe that's possible.

 

If this particular site wasn't an hour away with no internet hookup, I could do that. Not worth going out of my way to prove a point in an internet debate, though

[ak357 0]

cool guy

 

When designing a video analytics algorithm (say License Plate Recognition), there are certain pre-requisites like the pixel height of the License Plate text in the image, quality of the image etc which have to be taken into consideration for accurate recognition. With 1.3MP analog cameras, it might work best if you can focus the cameras for individual lanes. Just because a human can guess the text license plate from a 1.3MP camera covering 3 lanes (which I highly doubt), doesn't mean that the computer can recognize it too.

 

If you can upload a sample image from the 1.3 MP analog camera covering 3 lanes of traffic and still read the text on License Plate, your arguments can be taken seriously.

-----------------------------------------------------------

 

What is your opinion about number of horizontal pixels u need in order to capture lic plate for video analytics ?

Thx

 

would be nice if Soundy and jhonovich provide their opinion too

Thx

[frank3 0]

What is your opinion about number of horizontal pixels u need in order to capture lic plate for video analytics ?

Thx

 

From what I've seen and tried, mainstream license plate analytics require characters that are at least 50 pixels high to achieve decent accuracy, so as long as a character is 1/10th the height of a VGA camera, it should work. With 1.3 megapixel you should be able to have the complete car in your field of view, with a character being 1/20th the height of the picture. Thus I also doubt such a system would be sufficient to detect 3 lanes in an automated fashion, since a full license plate at that resolution would cover approximately 1/5th (20%) of the image width. The best license plate analytics module I've tried could detect characters as small as 10 pixels wide by 15 pixels high, but that was with a camera factory calibrated for that purpose, not a generic IP cam. Even with that kind of accuracy, a single camera was limited to two lanes or else sensitivity would drop.

 

I'd be curious to hear about your experiences and opinions too. I only tried 3 license plate systems so far.

[4MAX 0]

The best camera value I have used lately is the Samsung SDC415-NA. 550TV Lines and .002lux. OSD setup and works with 24vac or 12vdc.

 

As far as the DVR, Geovision has the ability to capture faces but only goes to 16 cameras. Don't know of any others that do that.

[Soundy 1]

That do what, only go to 16 cameras? Lots of them do that.

[jhonovich 0]

I believe 4MAX is referring to capturing faces. There are only a handful of companies that do that.