The Value of Video Analytics: Know More. Watch Less. Work smarter.
Did you know that most of us have been involved with video analytics for decades without knowing it? Video motion detection is the earliest – and most basic – form of video analytics.
In a typical scenario, an older analog camera would record an object in motion in its narrow field of view. It would perform a very basic “analysis” of the video, indicating some kind of spatial violation. Of course, with no discriminatory capabilities, the “violator” might be a leaf, a dog or some other non-threatening entity. Only by physically pouring through endless frames on premise could one establish the validity of the threat. A case of knowing too little, too late.
While this basic form of video motion detection remains one of the most widespread applications, we’ve come a long way. Currently, sophisticated video analytics is enabling us to see more, watch less, yet know more. And video analytics is paying off. The number of costly false alarms has decreased. Threats are being detected earlier, heading off potential lawsuits. Businesses are learning more about streamlining operations and attracting customers.
Enabling the Evolution of Video Analytics
There are several key contributors to the rise of video:
The need: Private companies, including big box retailers, are looking for ways to protect customers, staff and inventories. Law enforcement agencies, fire and rescue departments and other organizations that safeguard the public are being asked to do more with less. Video analytics can help them respond appropriately to trigger events. Savvy marketers are realizing the potential of excavating this goldmine of consumer information.
The technology: New applications are now available that make video security systems smarter and more streamlined than ever. The more analysis required, the more sophisticated the calculations. But the rule is, the application should dictate the technology. In some cases, less sophisticated “edge” devices (built into the cameras) are sufficient. But if more sophisticated analysis is required (way beyond distinguishing cat from a person on a roof), then software “embedded” in a PC or server is required.
The convergence of technologies enables motion detection to combine with access control, alarm systems and high-definition video to provide more – and better – information in realtime.
The network: IP (Internet Protocol) video is opening up the network to any business, any organization or anyone with access to the Internet. IP video incorporates robust capabilities that can carry companies into the future. More bandwidth empowers the IP network to carry more data wherever it needs to go for monitoring, analysis and response. Experts agree that IP will soon be the standard for video security, as older analog devices are replaced by digital versions.
The support: Companies are looking at the option of outsourcing some or all of their video, depending on their own expertise, available staffing and other cost of ownership factors, including bandwidth and integrator services costs.
More for the “discriminating” business
With the infrastructure in place and the expertise to use it, video analytics continues to evolve from the most basic application of motion detection to higher levels of discrimination.
Tripwire analysis can establish intrusion parameters. For example, if someone or something penetrates a security fence, an event notification can be triggered. If, someone leaps over a counter into a bank teller’s space, an event can be triggered. Taking it a step further can reveal how many people or objects (e.g., vehicles) are crossing a restricted boundary within a certain timeframe. This certainly has implications for both commercial security and Homeland Security initiatives.
Directional analysis is the ability to distinguish behavior by assigning specific values (low to high) to areas within a camera’s field of view. Just think of identifying the location, size and spread of a forest fire.
Objects-left-behind-or-removed analysis recognizes that a threat can come from an inanimate object, such as an unattended bag or package in an airport terminal. On the other hand, the unauthorized removal of public or corporate property would trigger the dispatch of a security guard or investigator.
Behavioral analysis is now being used for both security and marketing purposes. A vehicle in a shopping center parking lot circles endlessly, signaling a possible security situation. A person wanders aimlessly in that same parking lot, posing a potential threat to others or him/herself – and potential liability for management. Retail marketers are realizing tremendous benefit of analyzing store traffic patterns and the impact of product placement on product sales.
Removing the human equation through automation
Some might say that video analytics removes the human element from the surveillance process. To some extent it does. It removes the tedium involved in having one or more sets of eyes on a monitor for an extended period of time. In fact, the automation of video analytics allows the insertion of human judgment at the most critical time in the surveillance process. Now that’s technology doing what it was designed to do!