It seems to me that most Security Camera footage is so bad, not showing even the simplest of facial features clearly, that it’s hardly worth having the cameras at all.
Isn’t it about time there was an increase in camera quality, and in recording resolution, to actually make their existence worth it?
Eh. I think I put this in the wrong forum.
Until a mod wanders by and moves it I can have a go at an answer …
When many security systems are put in, they are on videotape, and the vid tapes get recycled time after time after time so the tape itself degrades. And again, many times the video cameras themselves are fairly cheap.
Hopefully with the drop in price of vid components and the improvement of imaging equipment and digital storage, the footage will be improved.
But then again, frequently the shots are compromised by where they have to place the cameras, giving less than optimal images because of the exaggerated camera angles.
There are good quality security cameras; they’re just expensive. I watched a TV show a few years back where some securty expert was talking about store owners complaining about the useless footage they got of criminals blurred or pixleated faces. He said that they, like people in any other business, try to go as cheap as they can on security and basically get what they pay for. You want a couple $200 cameras and a cheap VCR doing the recording, or do you want to spend $10,000 on a proper system to help protect your valuable twinkies and potato chips from shoplifting teenagers?
That’s another thing, I suspect security camera systems suffer from bloated retail prices.
I can’t figure that a few HD cameras + a PC would equal $10,000. Maybe half that. Not even.
Hmmm, maybe I should start my own business. Give the options for low price HD cameras, some basic wiring, and a quad core PC that streams the video to an off site server in case the bad guys get any funny ideas. Should be affordable.
Now who can I talk to about a business loan…
There’s more to it than just the system. I used to be an estimator for a company that sold and installed surveillance systems (in addition to its primary business of building and installing automated gate systems). This was a few years ago, BTW. As stated, there are cheap cameras and expensive cameras. A good camera would cost a couple-hundred to a few hundred dollars. How many cameras do you want? One? Four? Nine? Mounts have to be purchased, and options include weather housings, heaters, and such. Do you want a weatherproof housing on a pan-and-tilt mount with zoom capabilities that can be controlled remotely over the Internet? That costs. Cables have to be run and the hardware has to be mounted, and people have to be paid to do that.
Yeah, I said what I said tongue in cheek. It’s definitely going to be expensive to get high end video and multifunction cameras that might actually produce something that might help law enforcement.
I read an interview with some one in the banking industry once. His justification for this very question was: “On average a would be robber only gets away with X amount of cash. A security system with high def video would cost about three times as much. So it’s not cost effective.”
Which seems to me to be the point. If you don’t have good enough imagery, why have security at all? It can’t be just for show (well, not all the time).
The way to go, is for certain places to invest in the best technology available, make their results and effectiveness widely known so their deterrent value is maximised, then pretend to install them everywhere.
Moving thread from IMHO to General Questions.
One of the other issues that affects image quality is duration. I have a DVR-based security system here at my carwash, with 8 cameras and I can access it over the internet.
We have our DVR set to keep about a month’s worth of images before it starts to record over them. The shorter amount of time you ask the DVR to store information, the (slightly) better the image quality becomes, and by that it generally will have faster and faster frame rates the fewer and fewer amount of days you tell the unit to store in memory.
Of course, having it only keep a day or two in memory isn’t practical for us, so we keep it at about a month. This system has paid for itself many times over in debunking fraudulent damage claims by my customers.
I worked in Surveillance in a Las Vegas casino for a couple of years. I suggest that the lack of image quality is not the cameras but rather the fact that the tapes are running very slowly and are reused many times, basically until they wear out.
The cameras could zoom in to the point where you could literally count the hairs on someone’s knuckles while watching the live monitor. If a new tape happened to be in the VCR the recording was also pretty clear but if the tape had been in use for a long time the image was pretty grainy.
Tapes run at slow speed … a tape records an eight hour shift … tapes were reused every seven days. In the two years I was there we did a mass replacement of all the tapes only one time. A few tapes a day would get replaced when they broke or were removed from the cycle because they contained an incident deemed worth saving.
Why are the tapes used for so long? We ran 107 VCRs times 3 shifts per day times 7 days per week … makes for a lot of tapes. The decision is made seeking a balance between economy and usefulness.
The cameras at my place are very good indeed, good enough to be used as evidence in a prosecution, and thats images taken a couple hundred yards away.
There are aslo a number of features that can ensure that is anything of interest if noted, the subject will automatically be tracked and passed from camera to camera as the subject moves around.
This is only a medium secure establishment, the high security places have better equipment.
I spend all day, every day, handling matters like this.
First, even fairly cheap CCTV cams can be made to give wonderful shots if you focus them properly and design an environment where you can mount them without having your subjects backlit by the sun.
Second, security VCRs are almost a dead technology and about to be REALLY dead. The best time-lapse recorder maker out there, GYYR, quit producing recorders 5 years ago. That being said, if maintained properly, VCRs CAN give more than adequate pictures. Maintained properly, however, means new heads on the VCR every other year, and cleanings by a competent technician.
Anyone want to guess how often business owners neglect to maintain their VCRs properly?
I’ll note that there are regulations in NYC that effectively ban banks, for instance, from reusing VCR tapes taken from ATM surveillance shots. That helps tremendously with the messed-up tape thing. Really, why was anyone recycling those? You have money for a damned ATM and not enough for 1 VHS tape per week?
Third, the surveillance world is moving to digital video recorders. If you buy any decent DVR, you can get video ranging from ‘bad’ to ‘amazing’ by selecting among various compression levels on offer.
Just buy adequate hard disk size from the vendors, and before a compression or ‘image quality’ setting is chosen, ask the installer to show you what your footage will look like at “Low”, “Medium”, “High” or “Super-High” resolution.
Run the recorder at each resolution, walk around your facility, make faces at the camera, walk back to the recorder and LOOK at the pictures. Keep cranking the resolution up 'til it makes you happy.
I say all of this to say that the technology to give you good surveillance certainly exists, and IS NOT terribly expensive. Even vintage surveillance equipment like older CCTV cams and time-lapse VHS decks can work nicely, IF INSTALLED AND MAINTAINED PROPERLY.
If I use a digital recording system (hard drive or whatever), is it possible to design the software so it discards any unchanged footage? I worked at an office where we had security cameras in the hallways. I imagine that 99% of the time, the image is unchanged, so there’s no need to waste storage on that footage. (This was a few years ago, and we used videotape, but the quality was poor and reviewing the footage was tedious.)
Don’t discount the idea of deterrence. We can argue back and forth about how effective a camera mount might be for deterring shoplifting, or more violent crime, but ISTM that there’s at least some value to that. It’s like those signs that homeowners put out when they get a security system, no matter how good or bad it might be - a certain percentage of potential thieves will bypass such a house, just because their perceived risk has just gone up.
The thing that I’ve found unforgivable is not the shoddy cameras, but the companies that do not replace their looped VHS tapes on a regular basis. At this point we’re no longer talking about the thousands of dollars for a good or mediocre camera set up we’re talking about say $20 a year for maintaining whatever system you bothered to install in the first place. The marginal cost of doing routine maintenance just isn’t that great.
On preview: I just read Turble’s post - the system being discussed there is much more involved than the ones I was thinking of. I was thinking the system I’d met while working in a gas and go shop: We had, I think, four cameras, and one VCR recording the take from all four cameras. Even granting the costs that Turble’s system would have for proper maintenance are orders of magnitude more than the system I’m describing, I’ll stand by my claim that the marginal costs for doing proper maintenance are still negligible compared to the costs of installing, and manning, the system.
It’s also worth noting that routine maintenance costs are often viewed on accounting sheets as a pure loss for no benefit. So a certain mindset of the cost-conscious types will look at them, and ask “We’ve already got this system. Why do we need to keep paying more for it, when the maintenance doesn’t do anything?” A little thought about real-world consequences would show the fallacies in this thinking, but that thought often doesn’t happen. Not least because the people cutting the maintenance funds are often not the ones who have to deal with the consequences if the equipment fails.
Motion detection (really just change in pixels, but they call it motion detection) is a common feature with digital systems. I have the system at work set up to only keep 5-seconds before and after something moves.
A far simpler solution would be to have the camera begin recording for a set time (perhaps half an hour) every time a motion sensor goes off.