Leveling the Field: Artificial Intelligence’s Impact on Video Surveillance

For decades, video surveillance has fallen short of the expectations of many security operators. The benefits of video surveillance and its impact on operations, safety and security are undeniable; however, when you go deeper, video motion technology still suffers from the dynamic environments of real life.

Between lighting, shadows, weather and unpredictable movements by people and vehicles, video surveillance systems have gotten large. A single operator must manage upwards of double or even triple-digit camera feeds for situational awareness, and many rely on the cameras themselves or the video surveillance system to bring events to their attention for human action. As more video surveillance systems are digitized and placed on the network, though, the next digital revolution will be the use of artificial intelligence (AI) architecture and techniques to improve the efficiency of the detection and classification of objects and reduce false alerts.

With deep learning and large language models, AI techniques and architecture for images and video have accelerated in the past two years. As more vendors apply these technologies to their solutions, it will level the playing field for everyone. AI techniques that improve object detection and classification will become the new baseline that every camera provides. This level-setting feature will lead vendors to look at other creative ways to gain valuable insights from their devices and solutions.

Integrating metadata will be vital in fusing data points from cameras, sensors, access control systems and building systems. The pandemic accelerated the need for many organizations to understand their space better, and positive use of metadata is already seen in innovative search capabilities. The ability to quickly search through large amounts of video data to ascertain what you are looking for will be another level-setting feature that will be standard in every video surveillance system. A shift from reactive to proactive will result from these AI techniques and architectures, but this is only one component of a more extensive system and process that – with a human still in the loop – will drive a more efficient system.

AI disruption is real and will be great for competition. The customer will benefit from all of these advances and innovations and will discover new use cases. A key step will be applying ethical considerations and complying with the laws and regulations that will be forthcoming in the next few years.

The views and opinions expressed in guest posts and/or profiles are those of the authors or sources and do not necessarily reflect the official policy or position of the Security Industry Association.

This article originally appeared in All Things AI, a newsletter presented by the SIA AI Advisory Board.