What Video Analytics Can Do
In this article
In today's camera and analytics market, there are many use cases and benefits for a solution that is scalable, secure and sustainable. Across all industry verticals, intelligent cameras and video management systems (VMS), or video analytics, have purpose and value.
As the first part in a series on intelligent cameras and video analytics, we are going to review the most popular horizontal use cases. But first, let's start with an outlined understanding and approach.
Long gone are the days of closed captioning television (CCTV). As the sun sets on legacy camera solutions, a new combination of several components enters: video analytics. In short, video analytics is a combination of a smart camera, image intelligence at the edge and rule-sets on a corresponding VMS — or embedded in the camera based on the functionality and capability of the VMS.
Starting with the 'end in mind,' a mature video analytics solution can provide you with improved efficiencies, significant cost savings, increased security and real-time notifications.
For example, do you want to be notified when a high value fixed asset has been moved? Are you interested in role-based access controls (RBAC) with facial recognition integration as a second layer of authentication for highly secure or protected areas? Are you interested in audio or decibel levels which provide notifications for large groups or areas with crowds?
In many cases, buildings or locations are re-purposed for a use they may not have been initially meant for. This can create 'dead-zones' within locations, and these areas should have little to no traffic. We can utilize smart cameras as sensors to alert when someone enters these restricted areas for any reason.
For example, within construction sites cameras and geo-fencing rules at the edge can be applied to ensure 'no-go' zones are not entered. If they are, an alert is sent to a local resource to notify them that an area that has been breached, ensuring safety standards are maintained.
Smart cameras can utilize people counting to improve safety in manufacturing settings. For example, if a machine always requires two people to operate it, smart cameras can be used to alert floor managers if a machine is being used with less than two people. Additionally, people counting rules can be applied within airports, providing insight for appropriate staff flow and count based on higher foot traffic in specific areas or terminals.
In retail settings, people counting capabilities can be used to determine store traffic, dwell times, bounce rates, conversion rates and can also be used to identify profile matches to a retailer's targeted customer demographic. People counting can also help to identify traffic flow within a given location, allowing retailers to make targeted, specific business decisions.
Smart cameras will provide you with all your traditional asset monitoring capabilities while also adding the ability to integrate artificial intelligence and machine learning to enhance what can be accomplished with video.
For example, ensuring a fixed asset remains in a specific location until it is needed prevents theft. Additionally, a mobile asset moving across a field of view or area in a specific pattern can be captured using video cameras, ensuring compliance of that mobile asset.
Imagine if school security or local law enforcement could be automatically notified when a person that should not be on school grounds is there. This can be accomplished by feeding machine learning photographs of all teachers, administrators and students to the camera and asking it to send an alert if somebody shows up that does not match one of those pictures. Law enforcement can be immediately notified, and they will also receive a picture of the person in question.
We can use blacklisting functionality to perform more accurate people counting by identifying employees and eliminating them from the count. If we take this a step further, using a hospital as an example, blacklisting can help prevent theft. If we can identify doctors and nurses that should be obtaining medications for patients, blacklisting can be implemented to send an alert to security or local law enforcement if somebody other than a doctor or nurse is attempting to obtain medication that they should not have access to.
As we examine all of the analytical use cases across every vertical, the ability to enhance safety and security while developing business outcomes is greatly enhanced. The same technology that provides a physical deterrent to crime just by being present can also be used to drive business outcomes and a return on investment.
By transitioning how we think about physical surveillance and security video feeds, we are able to begin to move into the next generation of what is possible.
In the next part of our series, we will discuss smart camera capabilities around video motion detection, video auto tracking (along with license plate recognition) and many more analytics use cases for smart cameras. Make sure to follow our Internet of Things topic to stay up to date with this series and more IoT content.