Here's a look at the different scenarios when edge-based video analytics works better than analytics that runs on the server.
Video analytics has become an incredibly useful tool for the security and operations teams of many businesses. With instant, precise alerts, actionable insights, and increased ease of managing data, video analytics has revolutionized the way surveillance systems are used.
The technological advancements have also given customers more options on where to use analytics. What once used to run on servers can now work on edge or cloud as well. Many customers may even be better off using a combination of both edge and cloud-based analytics.
But such advancements have left many customers asking which is the best place to use video analytics for their purposes. There are so many different applications now, and customizability is the biggest advantage of video analytics. This article explores when you should use video analytics on the camera/edge.
Also read: How to increase public trust in facial recognition
When to select analytics on edge
Traditionally, video analytics have been run on-prem. And that has required full-fledged servers with graphics cards, etc., which are generally expensive solutions. With better technology coming in, we see a trend of video analytics moving away from the premise and migrating towards the edge, with cameras getting more and more AI functionality.
-
When you need real-time updates
One of the most obvious ways to decide if you need to use edge-based analytics is by considering the purpose. Would you need real-time alerts using people counting? Would you want to track customer behavior and get instant notifications? Would you want intrusion detection alerts? These are some of the situations edge-based analytics work better.
"A simple way to put it is that if you want something out of the box, and you want something standalone, a camera-based edge solution might be sufficient," explains Amol Kulkarni, VP and Country Head of Dragonfruit AI. "Some of the most common use cases of edge-based analytics are intrusion detection, slip and fall detection, spillage detection in retail or logistics environments, etc. Anything that that needs the real-time alert should be on-premise."
-
When internet connectivity is a concern
Cloud-based analytics require internet access. If you don't have the required bandwidth for any reason, you may want to consider edge-based analytics or on-premise server analytics.
This issue relevant, especially when your site is in a remote location where internet connectivity
-
When you have too many cameras
You may also want to choose edge-based analytics if you have many cameras, and it doesn't make sense to upload all the video to the cloud. The number of cameras worldwide continues to increase at a rapid pace because of increased affordability, ease of access, and better integration methods.
However, you may not need to run footage from all your cameras through an analytics solution. Some areas just need video surveillance coverage that you can use to investigate an issue later. It makes more sense to use edge-based analytics on a few cameras placed in areas that require it.
-
When cybersecurity bothers you
The third reason is when you don't want your data leaving your premises. In such a scenario, you can either run the analytics solution on edge or servers located on-premises. Many analytic solutions providers choose public clouds like AWS.
To be fair, the cybersecurity measures that these providers take are pretty stringent. Arush Kakkar, Founder of Agrex AI, pointed out that they have never had a customer who decided not to use a public cloud for cybersecurity concerns. But, there is always an element of risk that some customers may not want to take.
Conclusion
There are several advantages to choosing edge-based analytics over those on the cloud. But this doesn't mean that modern surveillance systems can do without cloud analytics. For many customers, a combination of cloud and on-premise edge analytics would be the best choice.