How to Keep Campus Dorms Healthier and Safer

Improving Security Operations with Video Content Analytics

As the COVID-19 pandemic evolves, it is an open question whether colleges and universities will welcome students back to campus in the fall of 2020. This is an important time for security teams to review their methods and best practices to create a safe, healthy and secure campus environment, and do it in a smart, efficient way.

Dormitories are some of the most important assets to safeguard, and they present some challenges because they accommodate many students — both residents and visitors — day and night; they are never fully closed, always open 24/7. Public safety teams must protect the buildings, and, more importantly, protect the students who live in them. Dorm security involves much more than conducting routine fire drills or having a student security guard at the entrance to “buzz” ID-carrying students through the door.

Campus police must prevent theft, assault, vandalism, as well as hazards that could cause fire or flooding in a building. They also are often the first responders on the scene of medical emergencies. Their job is two-fold: prevent incidents from happening, and conduct investigations after incidents. And now, in the midst of the COVID pandemic, security teams may also be required to monitor the hallways and common areas of dormitories to make sure that students are complying with health safety mandates such as face masks and physical distancing.

The important role of video surveillance and analytics

To monitor situations in real-time, or review video footage after an incident, most colleges have hundreds of CCTV video cameras deployed. Unfortunately, there is hardly ever enough manpower to accurately and effectively monitor every camera in real time. And, after incidents occur it is too time-consuming and ineffective to manually sift through footage from various cameras to glean information.

That’s why some campuses have implemented video content analytics software, which can be leveraged for complex object extraction, recognition, classification, and indexing activities, thus making video searchable, actionable and quantifiable. This technology helps campus security teams accelerate investigations, control who is allowed into buildings (such as dormitories), and send real-time alerts to security teams as threatening or emergency situations unfold. Comprehensive search and filtering capabilities make it possible to review footage from multiple cameras in a matter of minutes rather than hours or days. Furthermore, video analytics systems aggregate video data over time, so campus administrators can get quantitative reports about patterns and trends, such as pedestrian or vehicle traffic. This technology has broad campus-wide applications, but this blog post will focus on how it can improve dormitory security.

Investigate incidents more quickly and effectively

Video analytics software enables operators to search footage based on object classifications, such as cars, trucks, buses, motorcycles, bicycles, women, men, children, and animals, as well as face and license plate recognition and appearance similarity. The speed, paths, direction, and dwell durations of objects can also be detected and filtered for tracking down persons of interest in an investigation, after an incident of theft, assault or other illegal behavior has occurred in a public area inside or outside a dormitory.

For example, if a student reports that his bicycle was stolen from a bike rack outside a dormitory, campus police can quickly review the video footage from the camera near the bike rack to understand what occurred and, possibly, identify who stole the bike and when. Furthermore, they can search video from multiple cameras across the campus for other appearances of the suspected thief. Using filters that describe the person’s clothing, appearance similarity capabilities or even face recognition, campus law enforcement can more effectively track the most recent appearance of the individual on campus, and react accordingly to recover the stolen property.

Improve Response Time and Situational Awareness with Real-time Alerts

Real-time video analytics-based alerting capabilities dramatically improve awareness of and response time to situations as they evolve. Alerts can be based on face and license plate recognition, vehicles and people of interest, people or object counts, proximity identification, and line-crossing. Notifications can be sent to alert operations or security staff so they can quickly assess and respond to situations. 

Face recognition alerts

It is standard security practice to allow only approved residents, staff and visitors into a dormitory. Now, “In the Wild” Face Recognition technology can be applied, to complement access control tools and entry-point key and lock system technology. By compiling watchlists of recognized dorm residents and approved staff, security can be alerted whenever unrecognized individuals are detected on the premises after hours. These notifications can help video surveillance operators assess in real-time whether action or closer monitoring is necessary. 

License plate recognition alerts

The same watchlist configuration can be applied for In the Wild License Plate Recognition, as well, for detecting suspicious vehicles near the dormitory or any campus facility. For example, if campus officials have a concern about previous incidents that involve particular vehicles, they can create a watchlist of those license plates, and the video analytics system can notify the surveillance system operators in real-time if that license plate appears in a video camera feed.

Dwell alerts

Video content analytics can detect if a person or object is dwelling for a particularly long time in a particular camera view, based on pre-set rules; the operator of the system can set a duration to detect abnormally long dwelling. For some university campuses, theft is a major problem, and, sometimes, loitering for an unusually long time can be an indication of intent to commit a crime. Dwelling can also indicate a medical emergency. In either case, by configuring dwell alerts university law enforcement help bring events that may require emergency response to the fore, and allow for proactive and quick response. 

People-counting alerts

Video analytics systems can also be configured to detect anomalous behavior or objects, once the system operators have established normal patterns. For example, a video analytics system can detect the increase or decrease of people in a pre-defined area or view. System operators can configure rule-based people-counting alerts if the number of people in a dormitory hallway or lobby exceeds a pre-determined threshold, so that security can then assess the situation and respond with additional staff to disperse the crowd if necessary – this is especially important for reopening during the COVID-19 pandemic to enable social and physical distancing. By proactively discovering crowding – coupled with proximity detection – surveillance operators can quickly intervene in scenarios where the ability to distance is threatened and, thus mitigate the risk of virus spread.

Building occupancy alerts

Similarly, it has become critical for universities to track and demonstrate the number of people in a particular building, such as a dormitory, at a given time. Video content analysis can track the number of times that a line is crossed at an entrance or exit, alert when the permitted occupancy thresholds are exceeded and track occupancy on an ongoing basis.

Measuring & mitigating public health risks on-campus

In addition to private rooms, dormitories have common areas such as kitchens, hallways and lounges, where it may be necessary to monitor and track compliance with face mask and physical distancing rules. Because video content analytics software can recognize and analyze objects and behaviors, intelligent video surveillance operators can gather anonymous statistics regarding how often and where people are wearing face masks, how frequently physical & social distancing is violated – using proximity identification – and general occupancy trends that can help security monitor and encourage compliance with safety regulations.

Campus security teams can derive much more value from their existing video surveillance networks by implementing complementary video content analytics software. When security teams improve situational awareness and accelerate investigations, they create a safer campus, and a smarter, more streamlined security operation. The end result is better security and smarter use of security resources, not only in and around dormitories, but campus-wide.