AI technology is gaining momentum and becoming more reliable each day. In the security industry, we are seeing a myriad of AI applications, and a wealth of opportunities: both for consumer-end surveillance and central station monitoring.
Sifting Through Footage
One of the downfalls of security footage is that most systems do not have someone constantly monitoring the live feed. It's too expensive, unrealistic, and actually unreliable. AI technology, however, is helping to bridge that gap in a big way.
We are seeing new features and capabilities in the AI space being recognized in the security industry all the time now. For example, Comcast cameras now filter through motion activity, based off people or vehicles in order to help customers easily find footage of individuals coming to the door, or even to catch package thieves.
On an even more robust scale, IC Realtime’s Ella utilizes AI to help users sift through video footage with a surprising amount of detail. Ella has the ability to recognize extremely specific things, and allows users to implement search queries for types of animals, individual car makes and models, men or women, and clothing color.
In this way, AI technology has the potential to save everyday consumers, as well as law enforcement officials, and a huge amount of time, simply by making it easier to find the surveillance videos they are looking for and bypassing irrelevant information.
“The end goal of autonomous AI is that you really don’t need a human monitoring videos anymore. They simply monitor the events and react to those events,” says Shawn Guan, CEO of Umbo Computer Vision.
Facial Recognition
Facial recognition is another technology we’re seeing make waves in the surveillance space. And one that has not come onto the scene without controversy.
Through facial recognition, cameras are able to pinpoint and track specific people or individual characteristics. The application for this technology (specific to the security industry) includes:
- Crime Prevention
- Targeted Advertising
- Locating Missing Persons
- Protecting Performers or Law Enforcement
- Controlling Access to Restricted Areas
Identifying “Unusual” Behavior
One particular area that many facial recognition experts are interested in advancing is the identification of “unusual” behavior. Even beyond recognizing specific people marked as threats, this could take the form of reading the body language of anyone who comes through the door of a bank with ill intentions, or students in the hallway that are about to fight.
Boulder AI is one example of a company exploring the possibilities of this kind of advanced analysis. They are currently working on a system designed to analyze suspicious behaviors of people at banks. They are making use of both old security camera footage as well as shooting their own video with actors, in order to train their system.
“We’re specifically looking for bad guys, and detecting the difference between a normal actor and someone acting out of bounds,” says Boulder founder, Darren Odom.
An Application in Central Station Monitoring
Another place we could see AI enter the security industry is in central station monitoring. While most AI applications are currently being recognized on the front-end of user experience, there could be even more use for AI technology on the back-end as well.
Google Duplex is one example of an AI software that makes use of natural language in order to help customers interact with businesses. And while this technology is still in the early phase of implementation—currently only being used to book appointments at hair salons and make restaurant reservations—it’s not hard to imagine an application for this technology in businesses that rely heavily on human operators (i.e. central stations).
/the_potential_of_ai_in_central_station_monitoring.jpg)
