Threat Detected: How AI Video Surveillance Facilitates Active Monitoring

BlogBest Practices
Subscribe to the Singlewire Software Newsletter to receive the latest safety tips and trends delivered regularly to your inbox.

This guest blog post was written by Scott Keplinger, CEO of IntelliSee.

How Do You Know There is a Threat Detected?

Many organizations rely on video surveillance to help maintain a safe environment for their staff and patrons. While this can provide peace of mind, it also leads to passive monitoring as multiple cameras relay feeds that are impossible for human safety teams to all watch at once. This can mean organizations are missing out on the chance to know when there is a threat detected and has led to a heightened need for active monitoring solutions to help organizations better protect their people and anyone else who may be in their facilities.

To grasp the importance of active monitoring, it’s crucial to first understand reactive monitoring. A typical video management system stores hours and hours of recorded video that can be reviewed later as needed should an incident occur (if it’s not erased). This is reactive monitoring, meaning that surveillance is not being live-monitored and issues are not being flagged in real-time. In this case, recordings are useful in reaction to events, whether for forensics or to look back and gather clarifying information about what happened as the event took place. While this can be useful at times, wouldn’t it be more impactful to use that surveillance to intervene before an incident even occurs?

What is Active Monitoring?

Active monitoring allows you to proactively identify and intervene when there is a threat detected. For some large organizations, this involves an employee watching a video wall for long hours, scanning for something out of place or dangerous. Not only does this require thousands of paid hours a year—it’s simply not dependable.

Humans make mistakes, they get bored, and sometimes they get tunnel vision. How does this impact video surveillance? To start, the margin of error on manually monitored surveillance is predictably unpredictable. Infinite factors can impact a person’s attention to detail, ability to focus, or attention span each day. Not to mention, security departments are not immune to staffing shortages or unexpected absences. So, it’s often not a matter of if a human error will occur—but a matter of when.

That’s where deep-learning AI software comes into play. By overlaying AI software solutions with existing video surveillance systems, organizations can automate the task of monitoring video with a solution that identifies potential issues or threats the moment they appear for real-time situational awareness.

Be Prepared for Any Type of Threat Detected

Potential safety risks occur every day. Hazards range from spilled or leaking liquids—a seemingly harmless threat that can rack up thousands of dollars in slip-and-fall settlements or property damage—to weapons, people, or vehicles that pose a threat. Common risks can affect anyone, but catastrophic risks have the potential to impact everyone.

For example, gun violence is increasing but still extremely rare. classifies mass shootings as any instance where four or more people are shot or killed. By this measure, there have already been more than 600 mass shootings in 2023 alone, which is on pace with the 647 mass shootings in 2022.
AI software can’t eliminate gun violence, but it can transform your surveillance system from just a forensics resource to an immediate alert and prevention tool. With AI risk detection, the presence of a drawn weapon can be identified the instant it appears on camera and shared with stakeholders via customized alerts. Ideally, this all happens before the weapon is even used because the AI software has expedited situational awareness of the threat detected, saving precious time that can be the difference between an incident and a tragedy.

Don’t Forget About Everyday Risks (and Costs)

While gun violence is a real and horrific threat, organizations are statistically extremely unlikely to experience such events. What is more likely are common encounters like trespassing, medical incidents, and slip and falls. For example, the National Floor Safety Institute reports that 85 percent of workers’ compensation claims are attributable to slip and fall accidents on slick floors. Disabling injuries due to falls can cost businesses $250,000 to $300,000 per year.

AI overlaid on a video surveillance system can spot common threats like slipping hazards the moment they appear, allowing organizations to address the situation before someone has an incident. However, if someone does slip, AI can also detect a fallen person and alert personnel immediately so help can be provided as quickly as possible. This improves safety while also mitigating claim severity. It also allows organizations to bookmark and review footage before it is erased or becomes difficult to locate. This not only saves time, it can also protect against insurance fraud scams that might surface at a later date.

Different Threats—Same Tragic Results

In recent years, certain social media trends have racked up costly damages in vandalisms and thefts in schools, especially in bathrooms. While cameras aren’t permitted in bathroom settings, AI risk-mitigation can overlay cameras outside of bathrooms to identify loitering behaviors or flag when people are going in and out.

Parking lots and entry gates can be another high-risk area. On April 20, 2022, in Stockton, Calif., a high-school student was stabbed to death by a man trespassing on her school’s campus. The man entered campus through a parking lot gate and attacked moments later. Trespassers are just one of the risks AI can identify, and parameters can be put in place at any entrance, such as the parking lot gate, that should not have traffic at a given time or day.

AI combined with video surveillance works to combat non-violent tragedies as well. On January 29, 2019, windchills were 51 degrees below zero on a midwestern university campus when a student was accidentally trapped outside in the middle of the night. The doors had locked behind him, and nobody was present to let him back inside. Unfortunately, he was found deceased outside the next day. Surveillance video showed the student’s attempts to get help. The student’s mother has since filed a wrongful death lawsuit against the university. At another school, a student tragically took their own life beneath a stairwell. Despite the stairwell being in view of a camera, it was unmonitored, and the student was not discovered until it was too late.

Had a system been in place to alert someone that there was a time-sensitive threat detected, these tragedies could have been avoided for the friends, families, and administrations involved.

Unknown Needs for Medical Attention

Hospitals sometimes experience “dump and runs” where an individual is dumped at the ER entrance after they have sustained injuries. To avoid getting in trouble, the drivers leave without assisting the person inside or alerting medical staff to the situation. If there are no hospital personnel outside when this happens, that person is at risk of not receiving help in time, laying there until they’re noticed. Even if cameras are present, they are not useful unless actively monitored.

Solo workers represent another common yet similar situation. For example, an elementary school janitor was working the overnight shift when he began experiencing a medical emergency. After falling to the ground, he was unable to call for help, causing him to lay there for eight hours until he was eventually discovered. Janitors are not the only workers who face these risks, as many industries require people to work alone in isolated locations and workforce shortages often create unintended solo worker situations.

Once again, had an active monitoring system been in place to notify someone of a fallen person, the overnight janitor would have been aided much sooner and the chances of a worse outcome would be minimized.

Endless Risks. One Combined Solution There is a Threat Detected.

Knowing about a threat is half the battle, but the other half involves communicating and coordinating a response to minimize the impact and keep people safe and operations running smoothly. That’s why IntelliSee and Singlewire Software have partnered to integrate the AI video threat detection capabilities of IntelliSee with the mass notification and incident management solutions of InformaCast.

IntelliSee is an AI-risk mitigation platform protecting users against day-to-day risks, like slip and falls, while also protecting against potentially catastrophic threats, like active shooters. IntelliSee overlays with existing surveillance systems to actively monitor live camera feeds 24/7/365 to detect threats in real time. Once detected, instant alerts provide users with situational awareness, so they can verify that an issue has arisen. Using InformaCast, organizations can then manage the event by sending text and audio notifications to alert everyone, or specific groups, like safety teams, who can respond to the situation. InformaCast notifications can even include an image of a map with the location of the camera where there is a threat detected highlighted. IntelliSee’s AI detects drawn guns, slipping hazards such as leaks or spills, fallen persons, trespassing, loitering, vehicles, and other risks. This combined solution can be used in almost any environment where cameras are present, including K-12 schools and districts, colleges and universities, healthcare facilities, manufacturing plants, and offices.

Visit the IntelliSee website for more information about the benefits of active monitoring with AI video surveillance, and learn more about the partnership with IntelliSee and Singlewire Software here.