For much of the last 80 years, closed-circuit television systems were used to capture moments in time, allowing individuals to look back and review events after the fact. While there have been multiple advancements in video surveillance, condominiums have been slow to adopt them because of limited resources like time and money, lack of general awareness of new technologies, and hesitation to replace existing systems.

In essence, upgrading your corporation's CCTV system was never a "sexy" investment or a priority for most condominium managers or board members, unlike fun and more visible renovations to luxury amenities. Camera systems were simply a nagging requirement, installed or upgraded after an incident arose that was either inadequately captured with low-quality cameras or not captured at all.

For most of their existence, cameras have served as an effective tool for investigating incidents and providing evidence while offering little prevention.

This was supposed to change in the early 90s when video motion detection (VMD) was introduced. Users would identify areas they wanted to monitor and receive notification when there was a pixel change in that area. While VMD was considered a game-changer when it was released, it quickly became apparent that it lacked the intelligence to differentiate between actual security threats and benign occurrences such as lighting changes, animals, or foliage moving, leading to frequent false alarms. The time required for a security team to investigate these false alarms, coupled with the cost of implementation, rendered this technology practically unusable in condominiums.

Things changed in the early 2000s when basic video analytics were introduced. This was around the time properties began migrating from analog to digital technology, which worked better with motion/analytics. These analytics were an improvement to VMD in that they enabled cameras and recorders to recognize things like the height and width of objects, as well as speed, directionality, and repetition. This marked the beginning of a system's ability to differentiate between people, objects and vehicles and to send alerts to security teams based on preset conditions (such as people or vehicles moving in a particular direction). Though a significant development, the technology was still primitive and worked best with contrasting backgrounds. An individual in dark clothing moving in a white room or hallway would easily be identified; however, the analytics still struggled with complex environments and scenarios. It often mistook non-moving objects for vehicles and moving objects as people. While the frequency of false alarms reduced, they still occurred often, causing staff to ignore most alarms.

In the two decades since the introduction of video analytics, the technology has continued to develop and become more accurate. Still, the rate of adoption in condominiums has been slow due to cost concerns. Instead of spending to overhaul an entire video system, cost-sensitive condominiums have tended toward more affordable, piecemeal improvements to camera resolution, moving step-by-step from analog and low definition to high definition, ultra high definition, 4K, and beyond.

Fast forward to today, the rise of Artificial Intelligence technology has made intelligent video surveillance systems much more affordable and, thereby, more accessible to the average customer. It is no longer just high-end offices or large venues like arenas and stadiums with access to this advanced and user-friendly technology, but also residential buildings staffed by hourly, minimum-wage employees. This empowers condominium corporations to take a more proactive approach to risk, preventing issues before they arise and responding quickly to incidents in progress.

Gone are the days when condominiums needed their systems integrator to spend hours manually programming the system on specific rules and their exceptions. Instead, the corporation can type a security directive into an AI-powered machine (for example, creating an alert when someone checks two or more stairwell exit doors in a particular area), and the system will automatically configure the necessary logic and parameters. This reduces setup time and makes the system far more adaptable and easy to use, even for laypeople.

Video playback and incident investigation provide another example of change. Historically, looking into a security breach would require a guard or technician to sift through hours or even days of footage, all to find a particular frame showing a trespasser or an unauthorized individual damaging property. Even while performing a "smart search," which filters clips with motion in a specific area of the camera's field of view, users still need to review hundreds, if not thousands, of video clips.

With AI, users can ask it to find specific clips by describing what they show. This is especially useful when a resident reports a suspicious individual, like someone with a green hat on the property. Instead of looking through dozens of cameras and a large amount of footage, the concierge can ask the AI to "find a person with a green hat in the last 15 minutes". Within seconds, the AI will identify all relevant clips of individuals walking on the property with green hats within the specified time frame. Suppose suspicious activity is reported in real-time. In that case, there is now a possibility that security will be able to identify and locate a trespasser while they're still on the property, creating an opportunity to prevent mischief from happening at all.

AI of this calibre can track and identify a wide range of objects meeting certain criteria, such as clothing items in specific colours (jeans, white shoes, green hat, grey shirt, etc.), vehicles (box trucks, pickup trucks, cars, bicycles, including filtering down to the make/brand of the vehicle), license plates (allowing the search of whole or even partial plates), and even animals. If the AI doesn't find the right clip on the first try, the user can ask it to locate similar clips, helping it refine its search. This makes camera review quick, efficient and perhaps even enjoyable!

In some cases, AI combined with analytics even enables the proactive protection of people and property without needing intervention from a team member. Examples include triggering a strobe light to alert oncoming drivers about pedestrians crossing a busy driveway, detecting tailgaters following registered residents into a property, preventing daredevils from accessing the rooftop for dangerous stunts, or even warning an individual that they are in an unauthorized area. The system can actively instruct them to leave the property using flashing lights, LCD displays, speakers, and AI-generated speech. Then, if they don't comply, it can alert onsite staff or the authorities.

Consider the sharp increase in auto theft across the country. Most of us know that in a condominium, it's quite simple to follow a vehicle into an underground parking area without being detected. For a small investment, AI and analytics using a camera, strobe, and speaker installed at the garage door can detect unauthorized individuals passing through the overhead door on foot. With training, it can also easily differentiate authorized personnel, such as the condominium manager, security guard, and superintendent, from unauthorized trespassers, significantly reducing the likelihood of a false alarm.

As AI continues to reshape what is possible in the world of technology, it's becoming ever clearer that video surveillance is no longer just a reactive tool for cost-sensitive users - it's now a powerful, proactive solution accessible to the condominium market. It helps protect buildings and residents by deterring criminal activity while easing the burden on staff, allowing them to focus on other, more critical, community-building activities.

With the rapid advance of AI showing no signs of slowing down, the question is no longer if these technologies will become the norm in condominiums but when.

Michel Lauzon is a Director of Client Services with more than 15 years of experience in the security industry, rising through the ranks from Security Guard to Director within the Golden horseshoe Security Market.
smssecurity.ca

View PDF View Flipbook Back to Latest Issue