MAKING THE MOST OF CCTV MONITORING WITH AI

Like this one, our previous post considered Western Canada Security’s new video monitoring service. We spoke with the renowned Privacy Commissioner who has helped us better understand data ethics, public expectations, and legal requirements. This month, we introduce some of the AI analytics that your CCTV system can put to work, alongside a live stream of traditional security footage.

When you consider the convergence of AI and video monitoring, biometric surveillance might come first to mind. But that is out on the more extreme end of the AI spectrum (facial recognition technology; fingerprint or retina scanning). We find such intrusive methods unnecessary for most business contexts. But not to worry! Any manager feeling a little AI-curious in 2024 has a range of low-barrier advantages and insights to consider.

Real Intelligence

A quick peek behind-the-scenes: the range of smart-software products that converse with our video monitoring systems have turned out to be a relatively easy pitch. That’s because they leverage existing video feeds and entail minimal privacy risk. Most of the advanced tools under discussion do not recognize or store personal identities. But they do provide powerful, actionable insights based on robust data.

Let’s imagine that we have installed CCTV cameras in a local jewelry store. Its live video feed now brings second-by-second images into our local monitoring station. A massive amount of data is captured and transmitted, 24 hours a day. But much of that footage ends up having no purpose, because security incidents are few and far between. With our AI-enabled add-ons, however, we can harness security data for real business intelligence.

Heat mapping tools, for example, can learn precisely how customers move through the hypothetical jewelry store. Today’s smart-software products will analyze how people tend to travel between displays; where they spend the most time; and where they hit bottlenecks or spend too long deciding on just the right bracelet. Managers and AI can then collaborate on improvements to layout, product placement, and customer engagement. Data insights around busy and slow periods will fine tune decisions around staff schedules and security coverage. Over the medium to long-term, better in-store marketing and staffing efficiency is well worth the cost of obtaining the precise information needed to make improvements.

Strategic management has been ongoing since day one, sure. But the latest breakthrough is that this ongoing work can now lean on vast amounts of real-world data as well as sophisticated computer analysis.

AI can also monitor and analyze the store’s environment, including its energy use. In an instant, the manager accesses suggestions for minimizing costs and waste. Those changes boost sustainability efforts while serving as an early detection system for water leaks or problems with air quality.

If a dramatic or dangerous situation arises, our artificially intelligent monitoring systems will know immediately. And they will alert our team, who visually verifies the situation and dispatches appropriate resources. It still feels far in the future, the day when AI is trusted to dispatch Western Canada Security personnel. This technology serves us with data and suggestions, but we still rely on human intermediaries to take action on real-time security situations.

Traditional Security Functions

Unlike energy efficiency, or the sales insights gained from foot traffic, some of the AI tools available bring us back to our core offering: protection services.

Imagine one of our guards posted to as high risk a location as a jewelry store. This person has been expertly trained to watch for evidence that something is about to happen, not just to respond when trouble starts. This tactic involves watching body language, listening closely to voices, and even being on the lookout for concealed weapons. Today, AI can assist with such tasks (and remarkably well) for a minimal investment, especially if CCTV is already in place.

Because our conceptual jewelry store is surrounded by wall-to-wall glass, internal and externally mounted cameras have clear views of the floor area as well as the surrounding curbside and parking lot. If anyone spends more than a pre-set amount of time on the perimeter, managers are automatically notified to put eyes on the situation. An instant notification goes directly to Western Canada Security (thanks to a theoretical service contract) if any suspicious behaviour is detected.

Alarm response is another traditional security offering that works better with AI assistance. To illustrate, let’s give our imaginary jewelry store a fictional competitor. These folks have opted not to install security cameras. If their traditional security alarm goes off in the middle of the night, this second store’s security provider will have to dispatch a car and possibly multiple guards. If video monitoring were in place, AI tools could verify a false alarm with much, much greater certainty than someone watching remotely via video. That means reduced costs for more reliable performance.

In retail and beyond, experienced managers have an instinctive sense of traffic flows, busy periods, and the red flags that go up before incidents occur. But gut-level knowledge ultimately leads to “best guess” determinations and “hope for the best” solutions. With just one more push of the proverbial button, any client using our CCTV monitoring system can gain a transparent view onto their business operation and all the inevitable room for improvements.

https://www.securitywc.com/blog/