When you think of video as a security executive, you think of investigations, maybe searching through hours of videotape back in the day to find what you were looking for, or a massive wall of monitors with incoming feeds taken from all angles of a facility. Video today is invaluable to an organization — not only from this security monitoring perspective but from an overall intelligence-driven, decision-making standpoint. And this value is growing exponentially each day.
Video is the original sensor on the Internet of Things (IoT) and brings added value to an organization’s security strategy by allowing extensive connectivity along with the intelligence it provides. A video camera itself is arguably the richest data sensor mankind has ever created, bringing with it the information needed to be an enabler of more advanced technology, such as artificial intelligence (AI), deep learning, and analytics. Video is central to an organization’s security strategy — and without it, the level of oversight necessary for the safety and security of people and assets don’t exist.
But how can enterprise organizations implement video alongside other sensors to gather, dissect and analyze incoming data as part of this strategy? It begins with looking at the user, determining how video is used, displaying the information, and creating a unified view that brings it all together.
Look at the Video User
Who is the consumer of video surveillance technology? To answer this question is to delve deeper into an organization’s leadership framework, encompassing a number of stakeholders and departments that can gain significant amounts of information from incoming video meant for security purposes. That’s what makes it challenging. But these personas are integral to adopting a data-rich video surveillance strategy across an organization.
So, who are these users?
- Security: Primary users of surveillance, investments in video are usually driven by these leaders. But gone are the days when they are the only ones involved; now, a number of stakeholders weigh in.
- Business Operations: Streamlining how businesses actually DO business is one benefit to leveraging video data for multiple uses. In the retail environment, video and analytics can help ensure customer service expectations are being met; in airports, these same analytics can indicate whether to open a new lane for security processing, thus ensuring operational efficiency.
- Legal: Video for investigating slip-and-fall incidents or accidents has proven crucial for legal teams in protecting brand reputation.
- Marketing: Traffic patterns, dwell times, and other analytics-driven data is more important than ever for ensuring the flow of traffic through a store, but they can have some additional implications for testing and implementing various marketing strategies.
- Building Management: The nature of analytics has come far enough to control basic building elements, such as lighting, heating and cooling, based on occupancy levels and timing. This can have extensive implications for cost savings.
- IT: Today’s modern video solutions generally leverage IP cameras that reside on a network, requiring extensive involvement from the IT department in set-up and maintenance. They’ve become an integral part of the decision-making process.
One thing is for sure: Working together will allow us to collaborate to address the challenges that businesses have and help organizations and agencies improve the protection of their people, campuses, facilities, and spaces. But the driving force behind all of these users is the incoming video data that can serve all of these stakeholders and drive them forward in their own goals.
Video as a Data Enabler
As an enabler of technology, the potential that video and deep learning analytics provides is extremely substantial in detecting potential issues and events that require additional oversight. For example, within the healthcare space in a hospital corridor, if there is a prone patient or person on the floor, there can be an analytic in place that sends the info straight to a nurse’s station in the middle of the night letting them know that something is amiss and needs to be quickly addressed. The potential to combine deep learning and incoming video and actually use the information being collected has the potential to guide the use of video further than it’s ever been before: toward a more proactive, informed, and collaborative strategy.
The real possibility of advancing intelligence through deep learning and AI-driven technology applied to video is that we’re not going to be looking at the video until something has happened. This, however, might need a little more explanation. The goal of this high level of intelligence through video has the potential to be automated to the point that security operators will not be required to make the decisions necessary for response. Instead, the intelligence-driven next steps will be automatically communicated to various stakeholders — from on-site guards to local police/fire departments. Instead, when security leaders access video that corresponds to an incident, it will be because they want to see the incident for themselves. Yet, isn’t the automation, the ability to streamline response, and the instantaneous response to the goal of an overall, data-rich surveillance strategy? For almost any enterprise, the answer is yes.
This can be likened to a scene from one of my favorite 1980s movies: Back to the Future 3. In the scene, they’re talking about having cars in a bar and telling people that cars are being used and that now, people are only walking “for fun.” Everyone in this scene is making fun of the main character, but the same thing applies to video in security. As intelligence is placed via AI and deep learning technology, we’re not going to be looking at video to have all the answers anymore. Instead, having the video will be a luxury, not out of necessity. The data-rich nature of an organization (and how this data is analyzed and used) will enable decision-making and response in almost a secondary way.
In other words, analytics are currently used in the following way: a specific “rule” like motion detection triggers an alarm to a human operator, who views and verifies and decides on what to do with that information. This step may not be needed in the future as the combination of analytics, workflow and process, and integration may make the viewing of video more of a luxury. In essence, security leaders will know the situation was dealt with because the system took care of it.
And we aren’t a long way off from this. Significant advances that incorporate a large number of data points (with video being central to this) are already being implemented in large-scale enterprises that require immediate and proactive response to incoming threats as a matter of mission-critical protection.
Massive Potential for Data Display
As mentioned above, video is central to the incoming data across an organization, but there are dozens, if not hundreds and thousands, of other types of sensors out there being deployed across facilities, cities, and campuses — and they are all used for different purposes. Often, they’re used together. Sometimes they’re used separately. Regardless, the various sensors share one thing in common: they’re generating data.
To have a cohesive surveillance strategy, the data generated from these sensors must be displayed properly and, in the past, this has been difficult for a single manufacturer or platform to do. The mission to incorporate multiple streams of information into a single, unified overview for decision-making is driving many companies forward that utilize data-driven insights and deep learning. These displays not only look at video streams, but access control logs and door locks; external information, such as social media, weather, or news information; and other internal alarms to accomplish critical tasks such as mitigating potential risks, increasing manufacturing processes, or reviewing evidence during an investigation.
By leveraging a single-management platform, system management becomes much more accessible. An enterprise-class video management system allows stakeholders to easily see what they need to see when they want to view it. From entry-level surveillance set-ups to systems that support tens of thousands of cameras across multiple locations, these comprehensive software platforms can seamlessly integrate with various cameras, video intelligence software, and other networked devices to empower users to gain more awareness and make the most out of their investment in video.
Hinging on Interconnectivity
It’s hard to believe any of this is possible without focusing on interconnectivity and the ability to leverage a system seamlessly as needs change and morph. A prime example is the current landscape: in the midst of a pandemic with the very nature of how we work changes. As the country shifted to a more remote workforce, we’ve seen the need for organizations to adapt quickly and the security of facilities that were at low occupancy or even empty become paramount.
While analytics are an integral part of an organization’s overall security strategy, it’s beginning to look like they will be central to the goal of reopening businesses and “going back to work.” The ability to know occupancy levels, capacity monitoring, applying rules for maintaining social distancing, and limiting overcrowding will be in high demand across a number of sectors (namely, retail). That said, VMS and camera manufacturers that provide open-platform functionality will be poised to benefit from this demand. Integration with software companies that leverage AI and deep learning capabilities alongside video data can become a must in a “new normal.”
Central to a data-rich surveillance strategy is understanding the needs of the customer: communicating how you apply the technology forms the solution. We have to be able to understand what the customer needs before we can provide a technology that helps address the need. Buzzwords only go so far. Data-driven intelligence through deep learning and AI has the potential to solve some of the world’s biggest problems, so if all we do as an industry is to make better alarms, we’ve missed the boat. True intelligence will go further and is integral to the enterprise of the future.
About the author: Stuart Rawling is Vice President of Technology and Customer Engagement for Pelco. He currently serves as the Vice President of Technology and Customer Engagement for Pelco. He has held a variety of positions during his time at the company, including engineering and marketing leadership, consultant relations, and leading Pelco’s interoperability and openness strategy and execution. In his current role, Rawling works directly with customers to gain insights and feedback on how Pelco can assist with their business operations.
Rawling’s educational background focuses on technology and cryptography. He has a passion for all aspects of physical security and regularly speaks on cybersecurity issues and the challenges facing today’s physical security market. He is an active member of many technology and industry groups, including the Association of Computer Machinery, IEEE, the British Computer Society, and ASIS International. In 2016, he was elected to the ONVIF steering committee, and in 2019, he was nominated to the Security Industry Association executive council.
Original post: https://www.securityinfowatch.com/video-surveillance/article/21152240/how-to-implement-a-datarich-video-surveillance-strategy