Security experts have long looked to video analytics to perform functions that go beyond human capabilities. Whether it be detecting the slightest hint of motion or providing descriptions of multiple subjects and objects within a frame, video analytics generally do a good job of extracting key information from any given scene. However, this is where many video analytics solutions stop.
How do we move video analytics beyond descriptive to prescriptive? From gathering potential questions to providing solutions? The answer is more than just analytics — it is data intelligence.
The Second Wave of Video Analytics
The first wave of video analytics is marked by the ability to successfully extract both descriptive (men in blue shirts near red cars) and definitive (face, person, vehicle identities) attributes from available video streams. With this information, security professionals are better equipped to respond to security events (i.e., scan all surveillance feeds in a given area for red cars and men in blue shirts). Yet these analytics merely provide information while doing little to understand the context in which the information was gathered and disseminated. It is this context that matters and is a critical aspect of the second wave of video analytics.
Wave 2 video analytics work to correlate people, places and things as a means of better understanding both real-time events and forensic event history. From a security perspective, more event context equates to greater situational awareness, faster, more-appropriate event responses, and use cases well beyond security. In short, if the first wave was about information and data gathering, then the second wave is about dissecting this data to provide smarter recommendations and answers.
Data Intelligence and Pattern Analysis
Perhaps the strongest example of data intelligence in action is through correlation link analysis. Using machine learning and artificial intelligence (AI), correlation link analysis quickly establishes distinct patterns, matches events to those patterns, and finds anomalies where known patterns are violated. By understanding these patterns, security teams have a clearer picture of known people and their activities, drastically improving their situational awareness and investigative efforts.
In a public safety application, this kind of technology can be used to identify a person of interest as well as all of the other individuals with whom a person of interest has come in contact. Using a unique visual fingerprint, correlation link analysis does all the work, reviewing data from potentially thousands of video inputs to quickly identify the path of the person of interest through a facility or area.
Using AI, the system then correlates who the potential associates of this person are, where they met, on what dates, and at what time. In this way, correlation link analysis helps security teams understand the larger “web” of activity of persons of interest related to issues such as trafficking, theft and other criminal activities.
Correlation link analysis can also be applied to more than just people. Many organizations need to manage the flow of goods and information in their facility and knowing who is handling what at all times is paramount. For example, hospitals must properly dispose of hazardous waste materials and police operations are tasked with maintaining proper chain of custody over evidence. In these industries, any violation of regulatory processes could lead to costly fines or legal action. Correlation link analysis helps security teams understand when and where specific items are being transferred, providing insight regarding potential violations.
Implications of Data Intelligence
Correlation link analysis is just one example of the data intelligence afforded by Wave 2 analytics solutions. Data intelligence is also capable of producing valuable business intelligence for organizations to use beyond security. Take for example a video analytic that creates a heat map of spaces within an office, gathering data on where employees and guests spend their time within a facility. This data can now be fed into space planning systems via an application programming interface (API) so that departments well beyond security can derive greater value and deliver real cost savings.
Data intelligence also improves efficiencies within security operations while potentially lowering associated labor costs. By applying machine learning to the ever-growing body of metadata captured from video, the system is forced to control its accuracy, learn from its own outputs and distinguish specific priorities. This is the ‘intelligence’ afforded by data intelligence. In turn, the burden on human operators is lessened as more of the context-gathering responsibilities are relegated to the system. Security operations can now redeploy human capital where it is most needed most, such as alarm response, crime deterrence or simply providing a reassuring physical presence throughout the facility.
Looking to the Future
As the second wave of video analytics takes center stage, look to see the applications of data intelligence continue to grow as security data coalesces. Metadata gathered from other sensor types, such as access control and infrared camera data, will soon combine with Wave 2 video analytics to provide entirely new levels of data intelligence. This data intelligence stands to meet the growing needs of security operations and beyond, benefiting all market sectors in terms of greater speed, effectiveness and cost efficiencies.
