You’ve seen it in movies dozens of times. A fat, lazy security guard is sitting in front of a bank of cameras but is distracted by baseball on a portable tv. The hero just strolls right across the field of view of a camera or does some sort of trickery and gets into the top secret area.
I always get incredibly irked with those scenes because I keep on thinking if the designers of the system had built in some simple motion detection algorithms and alerted the guard whenever there was motion on any of the screens, 99% of those scenes would have been impossible. How dumb is it to rely on human eyes to manually scan a bank of cameras when there is software to do it for you.
Is that really what it’s like in real life? I’ve never heard of a human monitored security system that incorporated motion detection like that. It seems like such an obvious idea.
I’ve never seen a security system IRL without motion and entry sensors. In real life, all the sensitive areas have sensors that simply detect motion and alert anyone at the control panel that that specific area has something in motion. Ditto on the entry sensors, which are mounted on doors, windows, safe / display case doors, etc. and detect when something’s been opened. This applies to both attended and unattended versions of the systems.
Most older security cameras are analog, and lack any physical hardware capable of performing motion detection. Even the early digital cameras didn’t have the hardware to do anything more than digitize and transmit the signal.
This is changing as security systems are starting to use networked IP cameras more and more. These can have built-in DSPs for compression algorithms, which make it easy to perform motion detection at the same time (since modern video compression is heavily based on transmitting only the parts of the frame that are changing). So what you describe is becoming a feature in new security systems.
Laziness and money.
The technology is out there to bother the hell out of security professionals any time that there is motion in a zone that should not have motion and allow them to survey and then respond to violated zones.
There ARE facilities that use EXACTLY the kind of thing Shalmanese uses.
I’m a certified installer for a company that has a surveillance rig with the following features:
If someone goes onto the camera, it will notify you. You can adjust the speed it’s looking for and the object size it’s looking for, along with other factors. “Notify” can mean audio or video alerts, or it can tie to a big blinking light or an alarm siren. (Heck, you could use an old-timey 2-foot-wide bank bell if you felt like it.)
If someone goes west on a camera that is expected to have eastbound traffic, etc, it’ll trigger. [Handy feature for an airport, although I hope they’d ONLY use it to supplement a checkpoint with multiple guards on it in that case.]
If a crowd gathers, it will indicate. It can count a certain mass of new objects showing up and not moving in and out of scene… the authorities can decide if it’s a family reunion or a mob with hangman’s nooses, pitch forks and torches.
If a car is abandoned in a no-parking zone for more than a designated time, it’ll indicate.
Same thing for luggage.
After the fact, you can go back through your recordings and do searches for the above, or for a number of other items.
You can generate a custom display showing where all of this action is going on, in addition to dynamically-displayed scenes in question.
I know the system in question is NOT in a monopoly position. There are several vendors that can do all of the above, and dozens that can do most.
The problem is, the systems cost money, and figuring out how to use them effectively takes the time of experienced men, which can be just as expensive as the systems themselves.
Not to mention, if you add a bunch of highly sensitive alarms to a large facility, you can wind up needing an extra guard just to investigate your false positives.
PS - A LOT of things you see on TV regarded security are inaccurate. Particularly with bank heists, it’s a LOT harder in real life than on TV.
Several years ago I helped install some inexpensive network cameras here at work. They had motion detection algorithms which could be used to trigger various actions (start filming, send an email to so-and-so, etc).
That’s not true (unless by older you mean 1960’s era).
In the 80’s I worked for an alarm company. Our dispatch center was wired more than a 911 center. We had several cameras pointing at the doors to the building and dispatch center. The signal went to a breakout box. One feed went to the 24 hour recorder. The other went to a motion detection box before it went to the monitor. The detection box looked for changes in pixels over a set time period and buzzed when a change was detected.
It worked great for all indoor cameras. The outside ones would sometimes trip at night from car headlights, bugs, wind moving the camera, etc.
Yeah, it’s tricky to set an outdoor cam for motion.
The newer systems let you mask out areas that you don’t care about… you can usually get what you want inside, but if you have a road near your cam, headlights WILL sometimes give false positives for motion. If the pole, mount or pendant your cam on is shaky, you really ought not use motion unless your system just uses it to speed up record and you have storage to waste.
That’s why they use FLIR where I work to monitor the parking lot. Expensive yes, but when you have one of a kind R&D prototypes sitting outside of the high security area and intermingled with employee cars overnight it’s a must have. Of course I didn’t know this until the guard was laughing at me one day for peeing on my buddies car after we came back late one dark night. (it did need washed)
You’ve just confirmed what I said. The camera itself didn’t have motion detection capabilities, and was probably a simple analog system. To do motion detection you had to have a separate piece of hardware to digitize the signal and look for changes. These days most security cameras are digital, and use compression algorithms on the signal, which makes motion detection much easier - the same hardware which does the compression can also do the motion detection. The trick now is training the system not to generate false triggers.
Digitize? No, those older motion detection cam systems used either passive infrared (PIR) or radar to detect motion. There wasn’t any need to digitize anything. Many of the cameras of the day DID have a PIR sensor built right in. So yes, some cameras of the '80s and early '90s eras did have the capability to respond to motion without any additional hardware.
Of course the camera itself didn’t have the ability. But it IS part of the system. Nitpiking like that means I could say security cameras of today don’t do a lick of good because the camera doesn’t have a monitor to watch the signal - you had to have a separate piece of hardware to see the image.
These days most security cameras are STILL analog NTSC with BNC connectors. Some convert the signal to digital at the camera for network/IP based runs, some use fiber to transmit the signal (if you need long runs). If you go to a high security install like a bank or a casino, they’re all using analog cameras and wire runs. Sure, there is more digital devices in the security office but mostly for recording and storage or used for remote monitoring (convert the signal to digital and sent it out a leased line or the Internet to another office).
That is quite incorrect. Like REALLY incorrect.
The older motion detection video didn’t have PIR’s, AIR’s, radar or anything fancy like that. Now, I’m not talking about a system that looks for motion and moves the camera towards it. I’m talking about a still camera, pointing at an area, that can tell when something moves in front of it.
The cameras in these systems were regular, everyday video cams.
If you want to understand how these systems were wired, read my post above. I think it’s pretty easy to follow.
The device itself was analog and about the size of a old 1200 baud modem (about the size of a VHS tape). It had two BNC connectors on the back, one in, one out. There was also a small phono jack for an external alarm speaker and another for 12v input.
The front had one power switch, one knob for “sensitivity”, one knob for alarm volume, a “silence” switch and a little light which would go on when the device tripped.
In essence, it watched video signals. When it detected a change in contrast in the signal, it would go off. If you turned the sensitivity all the way down it would only trip if it lost video feed.
The company I worked for did a bunch of high end installs for rich bastards who wanted “hollywood” style security. If you were a rich bastard in the Seattle area, we most likely put in the system.
We got all the neat, new, cutting edge toys. Lazer gates, card readers, bioreaders, wireless devices (considered new at the time), glassbreak and audio sensors and groovy video systems. We installed a motion detector with an audio playback system for Sir Mix A-lot. If you went on his property and tripped the device it would tell you you’ve been detected and security was on their way (in his voice).
With all the high-tech toys we got to play with I don’t ever remember seeing a camera with a PIR. I might be mistaken, so if you’ve got a cite I’d like to see it.
It doesn’t really make sense to combine the two because placement of a PIR (or AIR) would be different than a security camera. With a PIR you want to try and cover as much of the room as possible. Normally in a corner facing away from glass doors and windows. With a video feed you also want to cover as much area as possible but you WANT to point it at doors and windows.
Consider something like an office reception area (three regular walls and one wall with windows and a glass entry door - think of a strip mall retail space). You want the video over the receptions head pointing at the entry door. This way you can see the reception desk and everyone coming in. The PIR would be on the wall with the entry door (or the corner of the same wall). The PIR would trip if someone opens the door or comes through a window or came into the space from behind reception. If the video camera was a PIR, it would see through the glass door and trip everytime someone walked by.
Of course sometimes placement would be the same, but if you’re putting two devices in an area you might as well give them different viewpoints for better coverage.
I work for a security company, and we have a camera system that’s supposed to have very sophisticated detection software. It’s supposed to detect a person crawling very slowly, for example. Like an inch every 10 minutes, or some ridiculously slow movement like that. I’ve only seen it detect a car driving a road, it put a white box around it on the monitor, that followed the car until it was out of camera range. Pretty cool.
A lot of the newer technology as pointed out is very costly. Depending on what your protecting it often doesn’t justify the cost minus insurance recovery.
Also remember if the sensor detects motion the crook knows this too and can false alarm it so much that it renders it meaningless. After all if every 2 minutes it’s registering motion it’s meaningless.
Again look at cost effectiveness. For instance here’s a good example. I had my wallet stolen at a gym late at closing. Within the 20 minutes it took me to call the credit card in. $800 was charged at a store. The store HAS the person on tape. The store WON’T persue it because they have insurance. The police department said they don’t go after “petty” theives and if I wanted I could persue it as a civil matter. But why should I? I got the charges taken off my credit card.
So the only person that was out was the credit card and me because after I filed the chargeback they upped my rate on the credit card from4.9% to 19.9% (I cancelled the account immediately. It was Citibank).
So that is a great example why persue something with costly technology when in real life no one cares.
Look, you just ran into floor limits on petty fraud and a situation unrelated to the OP’s example of a bank getting hit for a big score.
That DOES NOT mean the security system doesn’t work.
The bank doesn’t want to send out a detective to spend 4 days worth of work getting enough evidence to launch civil prosecution against someone who is likely judgement-proof. And they don’t want to spend $2000 on a lawyer to do the prosecuting.
The cops would pursue petty crooks if they had the budget, but in a lot of places they’d have to ignore persons who physically endanger other people in order to pursue petty thieves. I don’t want to live in a town that has to neglect either, but I’d rather have petty thieves than violent criminals loose.
If the thief had made off with $8 mil instead of 8K, the bank and the cops would have been after the bad guy like white on rice, but in your example the bank chose losing $800 over losing $3600.
Other than my fuzzy memory, I ain’t got one. But, since you worked in the industry around that time, I’ll happily defer to your expertise. I’m pretty sure I remember some PIR cams from the mid to late '80s, but it’s a tough thing to search for, apparently. All I can find are pages about modern equipment.
As for the radar thing, there are still systems which use radar to detect motion and aid in camera tracking–but they seem to be relegated to high-end security systems. Anyway, since radar motion sensors came in to being sometime during the late '70s or early '80s (I remember the old pressure-sensitive doors at supermarkets and the like slowly disappearing and being replaced by small radar sensors above the doors), I see no reason why there couldn’t have been such a thing to trigger a camera back then. Though, I’ll admit I don’t recall ever seeing one.