IDIS Global: From Conventional Surveillance to AI Video

As users move from conventional video into the new, AI-powered world, one of the most helpful options currently available is the new generation of edge AI cameras. With the latest tech releases, these devices are becoming more powerful, more capable, and – not surprisingly – more popular in a wider range of settings.

By: Dennis Choi, General Manager, IDIS Middle East & North Africa

Dennis Choi, General Manager, IDIS Middle East & North Africa

We are witnessing huge opportunities to leverage AI video technology to strengthen security, drive efficiency, and unlock new productivity-boosting applications across businesses. By running upgraded analytics engines, the new generation of edge AI cameras supports faster investigation tools without the need to rely on server processing, thereby also reducing load on central servers, and adding a layer of resilience, with edge processing and detection capability continuing to function in the event of network or server outages. Among the most compelling advantages of the latest edge AI cameras are ease of adoption, scalability, and lower total cost of ownership (TCO).

Avoiding Pitfalls With a Phased Adoption of AI Video

Ease of adoption is an important consideration, and edge AI cameras enable organizations to deploy and leverage analytics in a controlled, location-by-location manner.

This means they can avoid potentially costly purchasing errors, disappointing results, or – even worse in the case of security – equipment performance failures and breaches. They can minimize cybersecurity risks; assess platform reliability; manage any regulatory or compliance requirements; and, of course, control that all-important cost metric, Total Cost of Ownership (TCO).

There may be other considerations as well. For security teams to use the new AI surveillance tools effectively, changes to operations, dispatch, and incident-handling workflows need to be planned and managed. There’s little point in sending automated alerts to security personnel on their phones if response protocols and training haven’t been updated to ensure they’re always received and acted on quickly.

Usability Designed-in

In some cases, edge AI video analytics may be adopted for wider business functions beyond security – for example, a camera set up to send alerts when a queue builds in a reception area, or to notify staff when a vehicle arrives at a loading bay. There are countless applications, but to ensure the best value, teams that will leverage the new analytics and whose capabilities will be strengthened by it need to be closely involved in the implementation.

Where the plan is to leverage intelligence gathering from AI video, for example, to improve workflows, inventory management, marketing, or customer service, then the designers of the video solution need to work hand-in-glove with the department heads who will be using that data. The process of adopting edge AI cameras and leveraging new AI video surveillance tools is more effective when it involves all the relevant stakeholders from an early stage.

The process of adopting edge AI cameras and leveraging new AI video surveillance tools is more effective when it involves all the relevant stakeholders from an early stage.

Automated Threat Detection

Predominantly, though, we are seeing edge cameras adopted to strengthen security, plug vulnerabilities at key points, and enable automated detection of potential threats. By providing more reliable alerts and eliminating false alarms, edge AI enables faster, more consistent security responses. It’s freeing up staff from labor-intensive monitoring tasks, allowing them to be deployed more effectively, and ensuring more consistent threat detection without human error or lapses in attention.

Thanks to the latest, enhanced edge AI camera capabilities, the force-multiplying functionality of AI video is becoming more widely available to more users, across settings from retail and hospitality to healthcare and logistics.

The latest cameras enable automated detection of a wider range of risk events, offer easier real-time video search features, and can be enhanced with other useful innovations, including more accurate live privacy masking. Essential functions, including crowd detection, abandoned and removed object detection, and fall detection, support more preventative and proactive security and safety responses.

Fall detection, for example, is being adopted to improve safety in healthcare and assisted-living settings. The same technology is being used to reduce the risk of harm from accidents in industrial and logistics settings, and to improve safety in public spaces.

Removed or abandoned object detection is being used to detect potential terrorist activity in public spaces, transit networks, and critical infrastructure settings; elsewhere the same detection capability is providing automated warnings if fire extinguishers are removed, if fire doors are blocked or left open, if emergency exit routes are obstructed, or if high value equipment or assets are moved – everything from medical equipment to vehicles and machinery.

AI Live Monitoring Tools

And surveillance using edge AI cameras is even more powerful when combined with the latest AI live monitoring search tools designed to overcome the challenge of identifying crucial scenes within vast amounts of video footage.

These provide the option of making any target object or person easier to find by automatically extracting the best representative images, cropping them, and presenting them to operators for selection. By simply clicking the cropped image, the operator is taken to the precise moment in the video stream when the object last appears (i.e., the moment their most recent activity was captured on camera).

Searchable Visual Moments

Each cropped image can also be indexed as a searchable visual moment, supporting faster identification. When each cropped image is classified into categories, such as people, faces, and vehicles, AI identification of object attributes also allows for filtered searches – such as searching for people by age and gender, clothing color, or whether they have accessories like glasses, hats, masks, or bags; or searching for vehicles by color or type.

Thanks to these developments, where it would once have been too expensive or impractical for human eyes to monitor multiple scenes 24/7, AI edge cameras now make this practical and affordable.