Video analytics is a hot topic at industry tradeshows and in high-level security discussions. However, the analytics technology of yesterday — also known as Wave 1 video analytics — can struggle to provide context for higher levels of video understanding. 

The security industry is entering a new era of video analytics technology — enter Wave 2 video analytics.

Fueled by recent advancements in artificial intelligence (AI) technology and powered by flexible model architectures, the second wave of video analytics unlocks new capabilities for a number of use cases.

Wave 1 vs. Wave 2

From birth, humans are conditioned to take in a variety of factors in order to make sense of the world around them. It helps to think of intelligent video analytics in this way too, given that AI is largely based on adapting human thinking for technological purposes. 

Based on this idea, Wave 1 video analytics work to extract both definitive and data from video feeds, much like babies beginning to identify the shapes and colors around them. Security professionals can use this information to respond to security events as they happen, however there is little to no context. And without context, the human operators responding to such events are left to fill in the gaps, introducing the opportunity for errors, inefficiencies and a general lack of situational awareness.

But as the human brain evolves, so do video analytics. If the first wave was about information and data gathering, then the second wave is about providing recommendations and answers. Wave 2 video analytics builds upon the information from Wave 1 analytics to correlate people, places and things as a means of better understanding events and event history. 

Pattern-of-life discoveries and context-rich alerting functions — hallmarks of Wave 2 solutions — work inherently alongside human operators to provide clarity by placing the burden of responsibility on the system itself. In short, analytics, like humans, have learned how to best apply past experiences, known patterns, available data at-hand, and perceived potential outcomes to receive clarity on a given situation. 

Wave 2 video analytics by sector

Public safety

Pattern-of-life discoveries are an example of the computing and understanding power that Wave 2 video analytics may bring. Public safety professionals can use analytics that recognize patterns to identify suspicious behavior in public buildings or events, giving security professionals the opportunity to respond prior to an incident. 

Using machine learning, correlation link analysis tools establish distinct patterns, match events to those patterns, and find anomalies where known patterns are violated using past and present video data. Correlation link analysis helps users understand not only the relationships between individuals, but also build out the larger “web” of activity of persons of interest related to issues such as trafficking, theft and other criminal activities. It is these “pattern of life” discoveries that can assist public safety teams in their mission to quickly identify persons of interest and their larger network. 

Healthcare

In every hospital there are machines, pumps and other lifesaving technologies that need to be carefully monitored and protected to ensure availability when needed. Wave 2 analytics can track and locate specific pieces of equipment or individuals without the use of physical tracking devices or facial recognition. To do so, Wave 2 analytics make use of approaches like synthetic training data and transfer learning to extract information about the scene at hand. 

First responders

Wave 1 video analytics are known to work well on fixed cameras, but their ability to derive information from other imaging solutions stops there. First responders, for example, make use of a variety of different camera types and video sources including body-worn cameras and infrared cameras. Wave 2 rectifies this issue by applying analytics across varying imaging solutions.

Wave 2 and the bigger security puzzle

The above industries and Wave 2 technologies outlined above serve as an example of what is possible under the second wave of video analytics. Organizations in many industries can take advantage of the benefits offered by such solutions: more-informed event responses, a reduction in false alarms, redeployment of human capital to mission critical tasks, purpose-built analytics, and solutions for non-traditional camera types.

It is important to note that these capabilities are just a piece of a larger security puzzle that sees the coalescing of all available security data. Look to see metadata gathered from other sensor types, such as physical access control systems and secured entry solutions, combined with Wave 2 video analytics to provide new levels of intelligence. With greater security data coalescence, higher levels of contextual awareness can be achieved for even more secure, smarter facilities.