Loitering can be a precursor to other more serious crimes.
Growing up playing youth sports, a familiar battle cry from my coaches was, “A good defense beats a good offense.” All of us players dreamt of hitting a homerun in the bottom of the ninth inning or shooting that game winning bucket in a close game, but our coaches exhorted us to play stifling defense to eliminate any opportunity to even need to pull out a tight victory. Their mantra was defense wins championships.
Most experts would agree that when it comes to security, a winning gameplan centers around a stout defense mixed in with great detection and deterrence capabilities. The name of the game is to make your business as unattractive a target for crime as possible. That means making your business tougher to breach, or improving your ability to stop crime in action, or better yet, by preventing it from happening at all. Yet, with advances in technology like artificial intelligence (AI), another strategic piece of the playbook is emerging—the ability to predict potential crime before it happens.
One of the ways this can be effectively applied towards businesses is when it comes to combatting loitering. While loitering isn’t always a crime, the act of loitering can often be a harbinger to criminal acts and other unwanted behaviors that can negatively impact businesses.
Loitering and vagrancy can create an uncomfortable environment for potential customers. Panhandlers and people lingering around businesses may intimidate or scare off shoppers who don’t want to interact or feel hassled. The result is that the alienated customers shop elsewhere and the business’ overall sales decline.
For areas where loitering is more prevalent and vagrancy or panhandling is part of the equation, sometimes the interactions can get a bit more aggressive and escalate. This can lead to unsettling or unpleasant exchanges for both customers and employees. For employees, there is an added layer of concern as they may be in the unenviable position of asking loiterers to leave which may inflame the situation and potentially put themselves in danger.
If the trend continues where loitering impacts the number of visitors that frequent a business, the immediate area and surrounding community begins to feel the effects as well. The foot traffic wanes, there are less watchful eyes on the streets, and the area becomes more deserted. That begins to hurt the fabric of the communities the businesses serve, as neighborhoods start to become increasingly desolate. Long term, if security concerns persist, residents may flee the area and look for more secure neighborhoods to relocate to.
Despite the lack of statistics, it is widely believed that loitering is a precursor to more serious criminal acts. Areas that experience an increase in loitering and loitering times are believed to experience a correlating uptick in crime numbers. What is indisputable is that as businesses suffer from less clientele and the location and neighborhood become less populated, the area is much more susceptible to crime or nefarious activity. Loiterers can feel more emboldened and turn to theft, vandalism, drugs, or other illicit or lewd activities.
However, with predictive technological AI advances, surveillance equipment has experienced a slightly nuanced shift. Where the once primary objective of security equipment systems was to watch and wait for a crime to be committed, now with AI, surveillance systems are starting to predict the likelihood that a crime could potentially happen based on analytical behaviors. This shift in approach can be extremely useful when it comes to loitering detection. Where surveillance is typically used in a watch and see reactive strategy, AI is now turning the tide to a more proactive approach.
Portable security cameras coupled with AI can be effective in focusing on high traffic areas or areas that need more attention. AI software can analyze video in real time and automatically notify security if someone enters or lingers by a specific area whether it be a door, window, or any type of restricted area. When someone is detected in this hot spot of surveillance, the security system can automatically trigger a verbal warning through a speaker or escalate it further to security personnel or authorities.
Machine analytics can monitor individuals hanging around a building or storefront and keep track of how long they have been in the area. If someone is detected lingering for an extended period of time, it can identify the person as someone of interest and indicate a need for heightened surveillance. The time frame in which loitering raises suspicion can be set manually or can be based on machine learning based on collected travel patterns of other pedestrians. For example, if surveillance cameras in banks gather that the average ATM transaction is completed in three minutes, but someone is detected waiting or hovering by an ATM for longer than three minutes, that would trigger a warning that someone was potentially acting suspiciously.
Just as loitering times can be used to trigger more scrutiny so can the number of individuals in an area. For businesses where organized retail crime or flash mob thefts are a problem, the detection of groups of people congregating by a business can be cause for concern. AI can identify the number of people clustered together in a group as well as discern types of clothing or accessories they may be wearing. If a group of individuals are detected wearing face coverings or traveling with duffel bags, that could alert the system that something is potentially amidst.
AI enabled security cameras can also analyze traffic flow for everything from vehicles driving in a parking lot to the foot traffic in front of storefronts or building entrances. When it comes to loitering, it can identify individuals who are walking at slower speeds or frequently starting and stopping relative to other people around them. It can also look out for individuals who speed up near a secured entrance to make sure they aren’t trying to gain access by slipping in behind someone else who has authorization to enter.
Much of the predictive capabilities AI draws from is in its ability to detect and isolate various behavioral anomalies. The capacity of what AI enabled surveillance systems can do is getting more refined as it collects more data, and the machine learning adapts to that data. It takes a proven security model—utilizing the strength of surveillance cameras and mobile security units—combined with an ultra-quick processing brain. And while crime is often unpredictable and unforeseen, AI enabled security systems are starting to flex their muscles by giving us a clearer picture and better odds of identifying potential dangers in advance.
.webp)
