Archive for November, 2011
Recently, I was called out to a customer site to help them put together a camera system. To attempt to narrow the sales pitch, they put together their own “RFP.” Its clear from the RFP they did this themselves and did not use a consultant.
In the fall of 2001, sixteen Vicon digital security cameras were installed at the roof line around the perimeter of our building. A Kollector server was installed to capture images. Fifteen of the cameras had a fixed focus and direction and one of them was the pan, tilt, zoom (PTZ) type.
After a couple of incidences where we reviewed the images captured by the server, we realized that the system did not provide the information we had hoped for. While we could see a vehicle entering the grounds and a person near the vehicle, the images were not large enough nor clear enough to positively identify the individuals through their facial features nor could we be certain of the vehicle’s make, model, or license number. At night, lights were the only thing visible – headlights, tail lights, street lights, and parking lot lights.
In addition, the images captured were not always continuous. We might see a vehicle in the driveway, then in a parking spot, and then a person several yards from the vehicle, but all the action in between was missing. If there were several persons and vehicles moving about in the parking lot at the same time, one could not always be sure of which person belonged to which vehicle.
As time passed, we chose not to maintain a system that was unreliable. The Kollector server is no longer working and so we are not able to determine if any of the sixteen cameras are still working or not. A number of years have passed since the system has had any attention and the owner has now expressed an interest in a replacement system. There is no expectation to use any of the current hardware. On the other hand, there is no objection to using it if it is in good working order and fits into an overall scheme.
Objectives of a new CCTV Security System
At the present time, the owner is interested in only exterior cameras.
The cameras must be able to positively identify persons and vehicles and vehicle license plates in the parking lots and around the building, day and night. It is quite likely that an onsite real-time demo will be requested. The image collection system must be able to provide continuous imaging.
The following shows a couple of aerial views of the building which is 560 feet by 660 feet by 24 feet high. Employee and visitor parking and entrances are on the east side of the building. Receiving docks, generator, and vehicle storage entrances are on the north side of the building. Shipping docks are in an alcove in the southwest corner of the building. While there are no designed traffic areas in much of the area on the west side of the building and half of the north side, cameras will be needed to monitor for any inappropriate activity – graffiti, trash dumping, drug use, etc.”
First off, this is more thought than many of our customers put in to a design prior to meeting. It’s obvious they have a system that they are not happy with. They know what they would like to do and they know what they don’t want.
Anyone reputable who has put together a system sees lots of flaws in their design.
1) Just to get them back to reality I suggested that it could easily be $100,000 + to “positively identify persons in … the parking lots and around the building, day and night“. This was a crucial start point for the customer because it forced them to put a value to their needs. We were able to eliminate that requirement.
2) We needed to put a value on “… cameras will be needed to monitor for any inappropriate activity – graffiti, trash dumping, drug use, etc.” With a standard camera system, you would come in on a Monday morning, check the video, and see that there had been drug use on your property. But what could be done about it? Without a perfect face, and suspects already in the system, there is very little police could do with this video. The better alternative would be to do remote video monitoring based on motion activated cameras. This could run $175 per month and up. This posed the question to the customer is it worth $200 per month to catch the occasional drug deal? This customer decided it was so infrequent (1 time in 6 years) that it was not worth the cost.
3) The customer required “continuous imaging.” Without properly defining this, integrators could submit proposals based on capturing 5 up to 30 fps, causing quite a disparity in storage needs. We looked at some side by side comparisons and decided that 8-10 FPS would be acceptable.
This helped the customer stay within their definitions and get a better idea of what to expect from competing bids. In all actuality, the customer changed their design requirements again to capture license plates only at the 4 entry points, use their existing 14 working exterior cameras for overviews AND add 9 internal cameras. Originally the customer did not consider interior cameras, however, based on proper explanation, they came to understand that the 9 internal cameras will pay for the whole system.