Artificial intelligence is poised to change the face of video surveillance

Now that AI is at its peak, how will it change video surveillance technology?

surveillance video cameras
Thinkstock

For decades, video surveillance has been used to detect and deter criminal behavior in homes, businesses, and public spaces. But when you look back at footage from just a few years ago, it’s astonishing how rudimentary and grainy it looks. Not only is today’s video quality better, but so is the software behind it.

The evolution of video surveillance technology

When you look at the video surveillance industry and all of the new and advanced technology that’s hitting the market, it’s hard not to get excited about where things are headed.

Whereas you once had to call out a team of specialists to come in and hardwire a security system in your home—cutting through drywall and running webs of wire—it’s now possible for the average homeowner to install smart cameras with very few resources.

Whereas businesses once had to employ teams of people to watch security cameras around the clock, it’s now possible to automate much of the monitoring and have a security officer notified when there’s suspicious behavior.

The advances in video surveillance technology over the last few years have been tremendous, but we’re only scratching the surface. Something else is coming down the pipeline and it’s about to knock the industry into a whole different stratosphere.

The role of AI in video surveillance

As discussed extensively in this column, artificial intelligence (AI) is impacting just about every aspect of life and business. The security industry is getting a lot of attention and we’re beginning to see what a world with “smart” cameras could look like.

Typically, surveillance cameras provide citizens/businesses/law enforcement with the ability to either monitor situations in real-time or go back and review footage from something that happened in the past. The integration of AI technology into advanced security cameras could allow users to actually monitor situations in real time and identify issues before they become problems.

“Surveillance systems that include video analytics analyze video footage in real-time and detect abnormal activities that could pose a threat to an organization’s security,” data scientist Dr. Mahesh Saptharishi explains. “Essentially, video analytics technology helps security software ‘learn’ what is normal so it can identify unusual, and potentially harmful, behavior that a human alone may miss.”

This is one of the key driving factors behind the push to bring AI and video surveillance together. The idea is that advanced software could supplement human judgment and provide for more accurate and safe surveillance. It’s not meant to replace human monitoring, but rather make the process more detailed and personalized.

There has always been the desire to integrate AI and video surveillance, but up until this point, it hasn’t been technologically possible (from a hardware perspective). Well, that’s changing.

“The new frontier for AI and machine learning is being able to deploy these technologies in real-world products. This means getting the power demand to a low enough level to be embedded right into the cameras themselves,” says Remi El-Ouazzane, VP of a tech company that’s deploying a new processing technology for select surveillance cameras.

This new processing technology will reportedly enable new capabilities like crowd density monitoring, facial recognition, stereoscopic vision, facial recognition, detection of illegally parked vehicles, and behavior analysis.

Behavior analysis is something that a lot of tech companies are researching. The idea is that advanced security cameras can recognize precursor patterns that are associated with crimes and other public safety issues.

The West Japan Railway company is a great example. After discovering that 60 percent of people hit by trains in Japan are intoxicated, they installed 46 security cameras that automatically search for and detect signs of intoxication. The AI system looks for people who are napping on benches, stumbling, falling, or standing motionless for long periods of time and notifies human attendees who can then check to see if the person is in danger of harming themselves.

Anytime the topic of video surveillance is brought up, there are questions and concerns from the public about privacy. Nobody wants to feel like Big Brother is constantly watching. And while it might not put your mind at ease, those behind these technologies assure people that these systems know when to stop collecting information, as well as when to avert eyes. For example, a camera in the home can be told to shut down when certain “intimate” situations arise.

“I don’t want to give you too many examples, because as a French man I could get in trouble,” El-Ouazzane jokes.

Changes are coming

The integration of AI technology into advanced camera systems is something that’s inevitable and exciting. While there are always potential risks that come with increased surveillance, the potential pros clearly outweigh the cons. It could be a few years before we see mass adoption, but it’s exciting to anticipate yet another positive application for artificial intelligence.

This article is published as part of the IDG Contributor Network. Want to Join?