Gun-spotting camera can take the next step realtime and alert authorities

Credit: CC0 Public Domain

Technology makes it possible to catch criminal theft, violence, and mass shootings as it happens, and for viewing after the fact. What about revving up the camera technology so that even responsive action can happen realtime?

Particularly, how can we create technology to lessen casualties and mayhem tied up in horrific events of ? That is a question Lisa Falcone liked to think about, and her motivation to provide answers fed into her look at the current surveillance market. Surveillance: big whoop.

Cameras basically just recording crime... but just think if they could also communicate alarms and curtail the mayhem. Lisa Falzone is now the Athena company CEO. Falzone created the system technology with her co-founder, Chris Ciabarra, said Renae Reints in Fortune.

The Athena Security AI-based system can also offer an interface with third-party systems to lock doors, freeze up elevators, and communicate with people. Casualties in turn could be prevented, even some lives could be saved.

VentureBeat defined them as "a startup that's developing a machine learning layer for in retailers, corporations, and schools."

According to the company's official announcement, this is a launch of a security camera system capable of "instantly and accurately recognizing an active shooter before they shoot," sending out an alert to law enforcers. Athena's system is set up in one US school, Archbishop Wood High School in Warminster, Pennsylvania.

After spotting a gun, for example, it sends an alert to the business owner or authorities. It can stream live video to law enforcement too. That's not all. Digit said, "Users of this system have the option of hooking the camera up to other third-party security systems so that doors can be locked and lifts halted."

Pricing? The business model is per subscription type. The price range is $25 to $100. The system will be available in three tiers: Professional, Enterprise, and Unlimited Professional. Locklear said that the tier including gun detection, lock and elevator integration and real-time access costs $100 per camera per month.

Prices also reflect duration of service. A $50 fee wins 30 days' recording, a $25 fee wins 7 days' recording, and $100 gets the owner 1 year recordings. All carry cloud access.

Fine, but how accurate is the gun-detecting AI? Government Technology had an answer: 99 percent. "Since the gun detection is so accurate, the company is working on including recognition of other dangerous scenarios, such as knives and fist fights," said the report.

The accuracy rate is not ignored as a talking point. "False positives can be a problem for computer vision systems," said Mallory Locklear in Engadget, "and that would be particularly troublesome for a security camera that has the ability to alert police directly."

Once Athena detects a threat, a monitoring service service validates the threat and only involves the police when needed, said the company site. Even if the system were not to get it right, in other words amounting to nothing more than a false alarm, the system could learn from that.

Falzone said in VentureBeat that notifications have a "true or false" selector, making it easy to flag mistakes, with every flag feeding back into the system, improving its accuracy.

The system technology capabilities also pick up the store worker's stress levels, according to the company site.

Meanwhile, Kyle Wiggers in VentureBeat dove into some technical details.

The system involves "a Nvidia 2080 RTX graphics card and computer vision to detect what Falzone calls 'dangerous objects'—specifically guns. (Future versions will spot knives, too.) The AI system ingests footage from off-the-shelf security cameras and looks for 'dangerous motion' at a rate of 30 frames per second, alerting school administrators (or the police) via an app when a threat's detected."

More information:

© 2018 Tech Xplore

Citation: Gun-spotting camera can take the next step realtime and alert authorities (2018, September 30) retrieved 1 March 2024 from
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.

Explore further

Saving pedestrian lives by using public cameras to communicate with smartphones


Feedback to editors