Date Added: April 05, 2021 02:01:06 PM

In an article posted at The Verge https://www.theverge.com/2018/1/23/16907238/artificial-intelligence-surveillance-cameras-security, author James Vincent described how an Idaho resident used AI to identify and count fish making the uphill journey over a dam he had installed. Because dams can impede the ability of fish to reproduce, environmental regulations often require a dam to have a “fish ladder,” a series ascending pools that allow travelers to get up and over the dam. In this case, visitors had to check in. The dam owner was required to record both the number and type of fish using his ladder.

For a while, a person would sit at the dam and visually count and identify the fish. Then the owner installed a video camera so the inspection could be made remotely. Later, the owner moved to a system that used AI to identify the types of fish moving up the ladder. Now the dam owner could satisfy government rules with no humans involved. That's where video surveillance is heading. Artificial Intelligence is fast becoming the secret sauce inside video monitoring by putting brains behind the cameras.

So, what is Artificial Intelligence? The term AI generally refers to applying human intelligence to computer programs. The companion concept of Machine Learning is one of the techniques machines use to achieve a level of AI.

Think about how our brains function when they attempt to make sense out of an array of similar data points. When we're driving, our eyes scan and identify a continuously changing landscape of  objects: cars, trucks, people, bicycles and so on. And it's not enough for our minds to determine that a car is a car. We want to know the type of car, it's color, what direction is traveling and how fast it's moving. 

Video cameras are being equipped with AI software that helps computers perform tasks at a similar level. Initially, those have been repetitive tasks, like counting people, fish or things or noting when a person or vehicle passes through a gate or crosses an invisible boundary. Machines can do those jobs better and more efficiently humans. With computers doing the drudgery, people are free to do something with the data that being collected. They can interpret it, respond to it, and interact with it.

Law enforcement organizations have been the early adopters of AI video systems. are using AI to identify people by reading video images of their faces. Police and intelligence agencies are using AI-enhanced systems that search hours of video to find people wearing specific clothing. Systems that can identify the make and model of a car can determine where that vehicle was last seen. 

AI is also finding its way into video surveillance systems for commercial enterprises. At a construction site, security cameras can identify a backhoe from a forklift and track the movements of each item. In the retail sector, store owners use surveillance cameras with analytics to spot shoplifters and alert security personnel when the system sees something suspicious. Those systems can also analyze data to determine hot spots where theft is most likely to occur.

The good news for private enterprises investigating AI technology is that it doesn't have to involve replacing existing hardware. The AI magic often takes place in the cloud and on high-end computers connected to it. Many if not most AI deployments can be done using video feeds collected by inexpensive cameras and computers.

To further explore AI applications in video surveillance, contact Camdog.