How automation gives your existing surveillance cameras new skills

Automation and video surveillance: A perfect fit

A single office building, municipal center, or school can have dozens or even hundreds of cameras running, covering areas from hallways to community mailboxes, and often with overlapping fields of view. For large enterprises, the number can be in the tens of thousands. There’s simply too much video being gathered to ever watch all of it: most of that video is simply never seen unless an incident demands… and that means a lot of value in always-on cameras is lost. Eagle Eye Automations — integrated with the Eagle Eye Cloud VMS —  is intended to change that by turning detected events into action.

Modern surveillance systems can use AI to monitor the images gathered by cameras, precisely so no person has to do the tiresome and mistake-prone task of watching a camera simply because it’s on. Even better, you can use that information to trigger predefined actions with automation, whether that means dispatching a guard on foot to investigate, playing a warning announcement via a nearby speaker, flagging that area for a closer look by a remote monitoring service, or directly summoning police. 

Taking actions beyond security

And while surveillance cameras are most often primarily security tools, on the lookout for theft, trespassing, vandalism, and other intentional acts, their increasing capabilities mean they’re much more than that as well. Instead of being used to spot security problems, cameras can also send a note of caution or trigger an audible reminder to immediately resolve safety problems, such as missing personal protective equipment at a construction site. An automation tied to a camera view can alert managers when a hotel’s pool is in use so they can be sure a lifeguard is on duty, or turn off the lights on an office floor automatically when the floor is unoccupied.     

Automations don’t have to be simple one-and-done responses, either. By configuring linked actions that should occur based on situations you anticipate, you can initiate a whole list of reactions. That’s especially valuable in planning for situations where individual responses may be difficult to coordinate.    

What does automation mean in video surveillance?

For surveillance, practical automation can be described at its simplest as ingesting video data and recognizing the content of that video, configuring actions, and then triggering appropriate actions based on that content. Let’s consider all three of these at one more level of detail:   

Using AI to analyze content 

Video from all the connected cameras is analyzed for what it contains and metadata is created categorizing those contents. Does the frame contain a vehicle, a person, or another object on which the system has been trained? Is it a red truck? Is the person wearing a hat or a backpack? This recognition and categorization process creates the same metadata that enables easy natural-language searching of the same video content. 

Designing action plans

Based on the conditions you anticipate, you can design rules to guide what the system should do in a range of circumstances. Your rules will depend on formalizing your ideas of what circumstances to watch for or alert to; listing everyone who should receive alerts as a result and over what channels; and determining what actions should take place besides alerts. 

Executing decisions that have been configured

Data gathering and rule creation are necessary, but the value in automation is actually performing actions. This can mean: 

• Alerting specific people by email, text, or other means (“ALERT: Main entrance opened off-hours”).
• Triggering a motion, such as opening a gate or raising motorized bollards.
• Initiating high-resolution recording for additional scrutiny.
• Alerting a third-party monitoring service to investigate a detected event. 

Taken together, these aspects of automation mean you can plan both for everyday events (for instance, by opening a gate when a known employee drives onto the lot) or unexpected ones (a burst pipe causes a potentially dangerous condition), so you can react swiftly and with forethought. 

Features of Eagle Eye Automations

On detecting a defined event, Eagle Eye Automations can do all of the following:

• Send messages via text, by email, or through a messaging platform like Slack.
• Trigger on-site actions, including sending recorded messages via speakers associated with speakers.
• Connect to leading professional monitoring services, including Immix and Evalink.
• Alert system administrators when a camera is offline or has been tampered with.
• Initiate further action through third-party software integrations via Eagle Eye’s open API, for unlimited possibilities.
• Log and display all actions performed through the Eagle Eye Cloud VMS interface.

Practical concerns for implementing automations

Automations can only be as useful as the rules that define them, and the quality of the data that those rules are acting on. Here are a few practical steps that can make Automation most effective.

Evaluate the video quality and placement of the cameras that feed your automations. Whether you’re scanning for cars in a drive-through or watching for a crossed property line, you must ensure cameras have sufficient resolution and a properly set angle to be effective.

To reduce the cost of third-party monitoring, minimize false positives with AI. False alarms triggered by wind or animal motion are contributing factors to the high cost of that monitoring; Eagle Eye’s AI can eliminate nearly all such false alarms. 

Carefully vet your automation rules to be sure they’re addressing the problems where an automated response makes the most sense, and that the right people are set up for notifications. An automated message to an employee who’s left the company or is on vacation could be worse than useless. 

What automation will bring

Eagle Eye Automations, and automation more generally, are already in use and saving both time, attention, and money. You can expect that automations will continue to evolve with 

• Rule creation using natural language, for easy creation of complex actions 
• Integrations of ever more sensors (from vibration to sound-based alerts to tilt sensors) that will broaden the use cases for automated action
• An ever-wider range of hardware built with remote integration in mind, from speakers to gates, thermostats, security robots, elevators, and more

Want to learn how Eagle Eye Automations can make life easier?

Timothy Lord

Timothy Lord has witnessed and written about IT security trends and the ongoing evolution of SaaS for more than 25 years.