Video surveillance compared
Why false alarms and monitors staring yesterday was: AI-based surveillance in focus
This morning, I was jolted out of sleep by an ear-splitting alarm. Far too early, far too loud, and definitely not from my phone alarm. The third time this year, and as it turned out, another false alarm on the neighboring property. The monitoring of buildings and premises is an important measure to protect property. There are three basic approaches: traditional video surveillance with an alarm system (like my neighbor’s, a production facility), person-based real-time monitoring (not everyone can or wants to afford it), and AI-based real-time monitoring (AI-what?). Since the third false alarm this morning had me standing bolt upright in bed, this is the perfect occasion to explain in more detail how these monitoring systems actually work, what they achieve, and above all, what costs to expect.
Functionality
Traditional video surveillance systems consist of conventional surveillance cameras that continuously record footage. These systems operate passively, meaning they only capture events without immediately reacting to them. Therefore, you must also have an alarm system installed so that an alarm is triggered in the event of security-relevant incidents. This method seems like the perfect deal at first glance: “It only costs me the installation, a bit of maintenance, and occasionally a small fee for the false alarm.” False alarm rate depending on the system and technology up to 90% lies, that means the motion detector might go off out of sheer joy at a curious fox or during a storm. But hey, if nobody breaks in, it’s the cheapest alternative!
How does the break-in proceed
But let’s assume someone really breaks in, the process would probably be as follows: The burglars overcome the fence of the factory premises, the motion detector triggers an alarm, and the camera records the break-in. So far so good. However, since the premises have often triggered false alarms in the past – for example, due to animals or strong winds – the alarm may not be taken seriously (I didn’t take it seriously this morning either), and the perpetrators have enough time to cause damage and make a leisurely getaway. In this case, the material loss is often higher than the security costs.
The costs
The next day, the arduous work begins: The police spend hours analyzing the video footage. If they are lucky, they find a clue. The clearance rate in Germany is about 15%, in Switzerland around 25%. The material loss often exceeds the security costs, and the disillusionment is great. A system that only records and remains passive may seem inexpensive at first glance, but in an emergency, you pay a high price
Person-based real-time video surveillance
Functionality
Real-time monitoring works by continuously observing live video footage to immediately detect and respond to security incidents, following the principle: “Big Brother is watching you!” You either have your own employees who do nothing all day but stare at monitors and hope that something finally happens, or you hire an external security service to do exactly that for you. In both cases, people sit in front of screens and watch countless camera feeds. They not only have to keep an eye on the premises but also on data protection, and they are essentially just waiting for that one moment when they can say: “There is the intruder!”
How does the break-in proceed
Let’s assume the burglars finally sneak onto the premises, and at this crucial moment, the surveillance personnel are distracted – either because they have been staring at monitors for hours, where nothing usually happens, or because the guard is getting a coffee to stay awake. The sheer monotony of such tasks can quickly lead to fatigue. Perhaps the monitoring is also only limited because it is not staffed around the clock for privacy reasons or cost reasons
But even when everything runs smoothly, other pitfalls lurk: humans are simply not designed to watch video feeds for hours on end, where rarely anything happens (except for TikTok and Instagram). Eventually, concentration and reaction speed just diminish, and real threats can easily be overlooked. Additionally, frequent false alarms – triggered by animals, weather, or other harmless factors – minimize the staff’s attention and delay their response time
The costs
In the end, the false alarm rate may be better and the monitoring may be active in real-time, but you pay a small fortune for it. Traditional real-time monitoring with an internal security service requires a large number of employees to ensure 24/7 operation. Each operator can usually monitor only about 30 cameras, which means that for larger facilities, multiple employees per shift are needed. Personnel costs add up quickly. Additionally, there are costs for setup – including monitors, communication equipment, and regular training. External security services offer the possibility of outsourcing the infrastructure, but they are by no means cheaper. Here, not only hourly fees apply, which vary depending on the region and service provider, but often also surcharges for emergency deployments. Additionally, technical and organizational factors hinder efficiency: External service providers therefore cannot always supervise multiple companies in parallel, as the maximum number of cameras that can be monitored per operator is limited.
AI-based real-time video surveillance
Functionality
Now let’s talk about the “new hero of surveillance technology”: AI-based real-time video surveillance with Vaidio. Vaidio analyzes image data around the clock and responds immediately to relevant events, without the need for personnel to stare at monitors for hours. It detects suspicious activities in real-time and reliably distinguishes between real threats and harmless events. False alarms? Almost zero. The false alarm rate with Vaidio is reduced by 99.9%. A storm that makes the tree sway, or the cat that scurries across the screen, are simply recognized for what they are: harmless. Vaidio is like a security superhero that never sleeps, never gets tired, and never loses focus. Whether day or night, the system reacts in a split second. But it can do more than just detect break-ins: smoke development? Vaidio immediately sounds the alarm. An employee falls? Our security superhero recognizes it immediately and calls for help. This system is a true all-rounder that can be used in a variety of ways.
And how does the break-in proceed?
The burglars climb over the fence of the factory premises. Vaidio immediately recognizes the human movements, and even balaclavas, ski masks, or weapons are reliably detected. The security personnel are immediately alerted – and unlike traditional surveillance, the staff are not dulled by hours of uneventful monitor-staring. Instead, they respond alertly and attentively, promptly verify the alarm notification, and notify the police. So while the burglars are still eagerly anticipating a big haul as they approach the building, the emergency call is already in progress, and the response teams are on their way. Burglars often need only 5 to 15 minutes to carry out their act – exactly the time the police need to arrive at the crime scene after an emergency call. Thanks to Vaidio’s quick detection and the immediate response of the security personnel, this time window is optimally utilized to alert the police in time, cut off the escape routes for the perpetrators, and catch them.
Regarding the costs
A major advantage of AI-powered video analysis with Vaidio is the significantly reduced operating costs. While traditional surveillance systems allow an operator to monitor only about 30 cameras simultaneously, AI technology enables the monitoring of up to 500 cameras by a single operator. This is because the AI only issues alerts when truly suspicious activities are detected. Therefore, the operator no longer has to stare at screens for hours to possibly discover something but is specifically notified when a situation needs to be checked. This allows the operator to perform other tasks and not be permanently tied to the surveillance monitors
The savings are enormous: Less staff means lower salary and ancillary wage costs, especially in 24/7 operations where multiple shifts are usually required. The costs for infrastructure such as control rooms, monitors, and communication systems are also significantly reduced. Companies can therefore save not only on personnel costs but also on technical equipment.
Further possibilities
It becomes even more efficient when companies use Synaedge’s security control center. Instead of building their own expensive infrastructure, several companies essentially share the modern control center technology. This significantly reduces the monthly costs, and yet the companies benefit from state-of-the-art monitoring
In addition to savings on personnel and infrastructure costs, Vaidio offers the opportunity to improve operational processes through automated analyses. Errors are detected early, security gaps are closed, and inventories are efficiently monitored. This not only maximizes operational safety but also contributes to higher productivity and quality
Conclusion
In comparison, it becomes clear that both traditional video surveillance and person-based real-time monitoring are inefficient and costly. Traditional systems cause high follow-up costs due to numerous false alarms and offer only passive monitoring, which often reacts too late in the event of break-ins. Person-based real-time monitoring requires hours of live surveillance, usually without significant incidents, leading to reduced attention and high costs.
The AI-based real-time video surveillance with Vaidio solves these problems: It almost completely reduces false alarms and allows a single operator to effectively monitor up to 500 cameras. This makes better use of resources, as the operator is only notified of real incidents. Additionally, unused video data is transformed into valuable insights through intelligent analysis, which further saves costs and optimizes processes
Sources
Anne-Katrin Michelmann
Date 24.09.2024