By Madhukar Swayambhu
The word surveillance comes from a French phrase for “watching over” (“sur” means “from above” and “veiller” means “to watch”) therefore the concept of Surveillance is often taken as consistent perpetual monitoring of a place / person / area / region. Till the time, it is a small area like a small store, it’s okay. But the moment the area becomes larger, like a large format retail or a mall or a campus or a city, it goes quite complex.
Limitations of the human eye
With the advent of modern medical science, we now know the human eye very well and we also understand the limitations of the human eye. For example, we know that a human eye doesn’t have a capability for;
1- Night vision.
3- Thermal image.
4- X-ray vision etc.
Apart from the above, we also have a limitation of being able to focus on one subject and pay attention to the second subject.
Another limitation that we have in our vision is called “change blindness”. Change blindness is a surprising perceptual phenomenon that occurs when a change in a visual stimulus is introduced and the observer does not notice it. For example, observers often fail to notice major differences introduced into an image while it flickers off and on again. People’s poor ability to detect changes has been argued to reflect fundamental limitations of human attention. Change blindness has become a highly researched topic and some have argued that it may have important practical implications in areas such as eyewitness testimony and distractions while driving.
A yet another limitation we have, is called “in-attentional blindness”. In-attentional blindness, also known as perceptual blindness, is a psychological lack of attention and is not associated with any vision defects or deficits. It may be further defined as the event in which an individual fails to recognize an unexpected stimulus that is in plain sight. The term was coined by Arien Mack and Irvin Rock in 1992 and was used as the title of their book of the same name, published by MIT press in 1998
Ideally, when we know that a human eye has so many limitations, which can be resolved by going for electronic surveillance, then we should’ve replaced all the surveillance across the globe from human to electronic…isn’t it?
Isn’t that the most logical transition?
Limitation of the present surveillance systems
But we don’t see it happening, for sure, we see a huge sales growth for the cameras in an around the market, but all that is in addition to the human surveillance, not to replace it. The BIG question is – WHY? When the latter is better than the former, why is it complimentary? Why not a replacement?
The answer is simple logic.
A camera is better than a human eye, as it overcomes many limitations of human eye from the surveillance point of view, but it doesn’t compete or surpasses the human brain.
Imagine if we have 5000 cameras installed in a city, like the recent promise of one of the political parties in their manifesto, what we are doing is generating 5000 hours of video every hour, which means 208 days of video for just 1 hour. Who will watch it? And what will it give to the agencies? Just forensic data, after the disaster has already happened, provided the surveillance system is safe after the disaster.
Is that what is needed? Of course – NO.
The art of Suspicion
So let’s understand what we have in human brain, due to which the surveillance is till date under direct human purview and due to which the electronic surveillance is unable to help human being to do better in spite of cameras better than human eye.
The simple answer is – the capability to suspect. The art of suspicion.
A human being can doubt a suspicious activity. A human being can isolate abnormal incident / deed / activity. A human being can detect an odd man out or question a shady / fishy / iffy act by another human / object / machine.
To delve deeper, we need to understand this art of suspicion. What makes us suspect?
Anything abnormal is an act of suspicion. Some examples of shady activities, which cause suspension could be;
· Man leaves an object and moves on / Loitering.
· Somebody starts moving in the wrong direction.
· Somebody stops moving in the wrong direction.
· A Person starts running, while others are walking normally.
· A Person stops running, seeing the others around walking normally.
· Man entering restricted area or just standing there for hours doing nothing.
· Man exiting restricted area and looking around if anyone’s watching.
· Somebody walking on the railway track / metro line / top of a tower.
Now those were general suspicions. There can be more, depending upon the specific area, let’s say a parking lot;
· A man parking a two wheeler in a four wheeler parking lot or a cycle in a two wheeler parking lot.
· A man checking out multiple vehicles in a parking lot.
· A man driving too fast in a parking lot.
· A man covering his face while entering or going out of the parking lot.
· A vehicle with a covered / blank / unreadable number plate.
· Vehicle parking in wrong place
· Vehicle entering restricted area
· Vehicle parked in restricted area
· Vehicle over-speeding
· Vehicle makes illegal U-turn
· Vehicle pulled off road
· Vehicle starts driving in the wrong direction
· Vehicle stops driving in the wrong direction
Then there could be some specific intelligences like;
· Face recognition.
· Linking face to the vehicle (number plate) / memory database.
· Linking face to an object like briefcase / bag and associating the same with the person carrying it.
Above examples makes human brain better for surveillance than a camera, as the above capabilities makes human being to suspect.
The New Age Surveillance
Now a camera / surveillance system can be helpful only when it’s able to detect the above mentioned fishy activities, raise a doubt or suspicion and raise an alarm for the security agency / policing firm and make them alert. Without this, it’ll just be recording hours and hours of video which no one will have time, energy or patience to watch.
So we now have the basic answer – a surveillance system is useful only when it has capability to suspect and raise an alarm over a basic suspicious activity, else it is useless. That’s the basic need.
But beyond that – what else is possible?
That’s the concept of adding artificial intelligence to the surveillance system, then it becomes specific and more precise. Complimentary to the security / guarding system. Some of the AI features could be;
1- Face recognition linked to the attendance system / access control.
2- Number plate reading linked to face recognition & matching it with the database.
3- Head count – number of people coming in and going out and raising an alarm if someone is left in, after the designated hours.
4- Vehicle count linked to the size and color.
5- Over speeding of vehicles.
6- Wrong parking of vehicles.
7- Raising an alarm on an accident on a highway.
8- Thermal filters – detecting and living form in the visual span.
9- X-Ray filters – detecting the possessions of a person (if permitted legally).
10- Or more specialized like detecting a gas leak.
11- Speed monitoring of moving vehicles.
12- A virtual trip wire between the cameras to detect any object / person crossing and raising and alarm due to the said activity.
13- Night vision.
14- Movement sensor.
15- Time keeping.
16- Object left unattended.
17- Traffic & traffic signal management.
18- Bird counting.
19- Vehicle counting with color registration.
20- Black or white list for face / vehicle / object and raise alarm for the same.
21- Raising an alarm on tampering of cameras.
Now if all those features can get integrated with the camera with the help of an artificial intelligence mechanism or software or hardware, the whole concept of electronic surveillance would start making sense for the security / monitoring agencies.
Ultimately the whole concept of technology is to make life more comfortable / easier, anything adding to the labor / work / botheration of human being is not technology – its error.
The Key ingredients of a Solution
Now let’s understand one more aspect of surveillance systems – for electronic surveillance. A surveillance system need 5S components which are;
1- Surveillance Camera
5- Switches (networking)
If anyone of the above is missing, the solution is incomplete. Therefore, it is important to understand the five elements before deciding for the surveillance solution and anyone vendor who doesn’t have the complete components is just giving you yet another problem and not a solution.
So be aware and buy wisely. Ultimately, it is not about the cost but the solution to the problem.