Features
The role of AI-based surveillance in improving emergency response times

For first responders, time is everything.

Regulators in Canada’s neighbouring country estimate that as many as 10,000 lives could be saved each year by reducing emergency response times by just one minute. Police, fire and ambulatory services all rely on technology to improve the efficiency and effectiveness of their workflow. Technology can play an important role in large events like parades, protests, sports events, and generally crowded areas, while also helping more isolated incidents local to neighbourhoods. Video surveillance has emerged as an essential tool for first responders, as it allows them to quickly understand a variety of emergency situations, whether they be medical emergencies, fires, or situations requiring police intervention.

Of course, video surveillance is nothing new, but the advancement of AI has brought about highly intelligent cameras capable of more than just streaming video, as they can now understand the footage they are transmitting, to a degree. With built-in AI analytics, cameras can generate metadata that is attached to objects or events within the camera’s field of view. This benefits a video operator in real time, as they can be shown the most relevant video footage at any given moment. The number of surveillance cameras far outnumbers their human operators, after all. It also aids review processes or investigations that follow an emergency situation. Investigators can filter footage of an area to timestamps where the camera recognized human presences, instead of trawling through hours of empty, irrelevant footage to find what is needed.

This article will explore how first responders are benefitting in real-time by the use of AI-analytics backed surveillance systems. This technology has the potential to revolutionize emergency response protocols by enabling faster response times to critical situations. It is, however, not without its limitations and ethical concerns, so those will also be addressed.

Real-time analysis of surveillance footage

The basis of the benefits of AI-enabled video surveillance systems stems from their ability to perform real-time analysis of surveillance footage. Video footage can be analyzed in real-time by machine learning algorithms, which can aid in detecting potential threats, tracking suspects, victims or missing persons, and providing heightened situational awareness to first responders.

For example, in the case of a missing person or a suspect on the run, AI-enabled surveillance cameras throughout a city can perform facial scans in real-time to try and detect matches. These matches can then be reviewed by operators individually, to ensure they are looking at the most likely matches instead of thousands of people in a given area. Analyzing this data in real-time means emergency personnel can immediately mobilize and respond to sightings of missing or dangerous individuals, affording more safety to them and the public.

Real-time analytics can benefit each branch of emergency services in different ways, but the goal is always the same: identify emergency situations as fast as possible so the proper response can be deployed.

Real-time analytics and emergency response times

When lives are on the line, every second counts. Any sort of advantage that emergency responders can gain could translate into multiple lives saved. Let’s look at a few situations where first responders would benefit from being backed up or informed by AI-enabled surveillance systems.

Law enforcement is perhaps the branch which can apply AI-enabled surveillance systems to the largest variety of situations, and thus has the potential to speed up response times for various emergencies. For example, surveillance cameras and sensors monitoring a particular area of a city can detect when certain sounds are heard, such as that of a gunshot. This information can then be immediately relayed to law enforcement personnel for them to handle the situation. The alternative is relying on eyewitnesses to see something or hear the gunshot and then alert emergency services. There can be a delay in the emergency response that is eliminated by AI-enabled surveillance systems recognizing the event in real-time. Licence plate recognition is another powerful tool, where an AI-backed system can identify licence plates belonging to vehicles that were stolen or involved in criminal acts. Sometimes, alerts can be issued before a situation arises. If, for example, an individual has trespassed into an area and spent a certain amount of time there (long enough to rule out a mistaken journey), then law enforcement can be automatically notified, in case the trespasser has malicious intentions.

Paramedics rely on getting to people as fast as they can, to rescue as many people as possible. AI-based surveillance systems can play a large role in facilitating faster responses to emergencies, through a few key functionalities. For example, a bustling city centre hosting an event during a hot summer’s day may have emergency personnel on standby to make sure citizens are staying safe in the sun. AI-enabled surveillance systems could monitor large crowds, detecting instances of people collapsing or showing other signs of heat stroke. First responders could be immediately notified, instead of relying on manually scanning the crowd or listening out for calls for assistance. On a larger scale, AI-enabled surveillance systems which span an entire smart city could analyze current traffic across large areas, providing the fastest possible route for ambulances to take to the scene of an emergency.

Firefighters also benefit from being on the scene as quickly as possible, with as complete a picture of the situation as possible. AI-based surveillance systems can assist in both regards, with instant alerts for when fires are visually detected, as well as thermal cameras which can monitor temperatures and alert emergency personnel when temperatures reach potentially dangerous levels, which indicate a fire may break out soon. Once a fire is detected, AI-based surveillance systems could swiftly identify, through past footage, the number of people within the structure.

Potential ethical concerns

According to a recent study, only 51 per cent of Americans would consider themselves comfortable living in a “smart city,” so this is something to consider when implementing AI-enabled surveillance systems. It could be argued that the discomfort people associate with smart city technology stems from a lack of understanding, but there are valid ethical concerns which must be addressed to make sure all parties understand the technology and how it is used.

Most ethical concerns revolve around privacy and a lack of transparency. A company using facial recognition to find missing people has a good end goal in mind, but people may not want their faces scanned and added to a database as they walk around in public. A middle ground must be reached for these technologies to experience widespread adoption, and they only become more effective the more widespread their use becomes. Transparency also needs to be handled with care, it’s important to let people know how the technology works, and its intended use, but divulging too much information could affect the privacy of certain individuals or grant malicious actors too much insight into a process meant to stop their behaviour.

However, these concerns can be addressed by using privacy-enhancing technologies, regular audits and reviews, and transparency in decision-making. It will just take time and trialing.

Conclusion

For first responders, AI-enabled video surveillance can be a powerful tool that shaves valuable seconds off average emergency response times. Whether it helps enhance the situational awareness of first responders, aids in tracking suspects or victims, or helps catch accidents before they happen, AI-enabled video surveillance has clear benefits for first responders, so long as ethical and privacy concerns are addressed and communicated to the public.

Sophie Laplante is the business development manager, cities, Canada with Axis Communications.