Page 18 - Security Today, March 2021

for a man with a blue shirt and shorts. With the embedded meta- data provided by an AI-based camera, the search yields results within seconds.
AI technology can even extract clues to behaviors like falling down or fighting by using human skeletal characteristics to classify how people are positioned. Being able to search through this em- bedded metadata on a VMS will require a plugin or API that can read the data. This capability will be available in Hanwha’s Wisenet WAVE VMS and plugins are also being developed for VMS sys- tems such as Genetec’s Security Center that make integration easy.
Applications for Business Operations – Going Beyond Security
With all of this collected data, there’s never been a better time to start thinking about video cameras in a broader sense. AI-based cameras will become important data collection sensors which go far beyond simply capturing video.
They can identify and count objects, display heat maps and ensure process and operational compliance. As a result, they will become an invaluable tool for business operations. Depending on the business, the value proposition for such data can be a game- changer worth many times the cost of the system. Cameras are already well accepted and commonplace, so the opportunities for them to evolve into unobtrusive, important data gathering tools for business and operations intelligence will only continue to grow.
Once seen as merely protecting the bottom line from loss, AI cameras can truly be seen as an enabling technology for revenue generation. They can be tools for operations and process measur- ing and metrics.
Bringing AI Data back from the Edge
With AI on the edge, the valuable events and other metadata gen- erated by cameras will need to be gathered from many endpoints and the data aggregated together, usually in server or cloud-based systems, to enable clear visualization of the trends and anomalies identified. This can be presented via a suite of basic dashboard and trend visualization tools for a variety of customers.
For customers with more sophisticated needs or with various other types of data to be used in analysis, the camera metadata can be accessed and combined with other data and processed by other platforms for sophisticated visualization and data mining. This allows technology partners to access the data we aggregate into their own charts, graphs and exception reports powered by specialized software companies they may already be using.
There are familiar use cases spanning multiple industries that require linking data from access control, intrusion, point of sale systems, staffing data, schedule data, weather data, or many other data sources. The potential for unified data to create comprehen- sive business solutions is substantial.
Future-Ready: Reducing Risk and Cost
in Migrating to AI in a Surveillance Environment As AI-based cameras start out with only limited market share, it’s important to make upgrading to the technology as easy as pos- sible for integrators. For this reason, Hanwha’s AI-based cameras will be based on magnetic camera modules that will quickly and easily snap in to any pre-existing Wisenet X series PLUS form factors. This will make migration from non-AI to AI as smooth and efficient as possible.
Server-based AI definitely has its place in many use cases. The greater compute resources available with server-based solutions will allow workflows not possible on camera hardware. However, because there are so many architectural advantages to edge-based computing on the camera, AI-based cameras are now poised to evolve the traditional security business, and the video endpoints it implements, into the IoT data-gathering business.
When talking to customers about AI, there is a clear desire for help with operations and processing needs. In many custom- ers’ minds, AI represents a limitless potential to enhance business operations and logistics.
The video security industry now has the opportunity to re- invent itself as cameras and supporting infrastructure become smart sensors capable of assisting in day-to-day business opera- tions and logistics. In addition to better serving traditional securi- ty detection use cases, AI-based cameras can gather information which directly impacts purposes beyond traditional security in- cluding revenue generation, operational efficiency and customer experience in unique and powerful ways. As such, they have the potential to transform our industry and open doors to new busi- ness opportunities like never before.
Ray Cooke is the vice president, Products, Solutions and Integra- tions, at Hanwha Techwin America.
WHAT IS AI AND HOW DOES IT SERVE VIDEO SURVEILLANCE?
AI is the technology that enables computer systems like cameras to perform tasks which normally require human intelligence.
Today we are working with a subset of AI called “machine learn- ing,” and within that category the technology for analyzing video that is ready to serve in systems is “deep learning.” Deep learning soft- ware attempts to mimic the activity in the neocortex, to interpret what the system “sees.”
In these models, an array of deep learning software algorithms are fed massive amounts of curated data so they can learn to recognize patterns in digital images. Just like humans, the algorithm learns by being shown many examples. In fact, for a deep learning algorithm to be truly accurate, it may need to be fed many hundreds of thousands or even millions of sample clips and images. And, if it generates a wrong result when making predictions, it needs to be told that too. Thus, it “learns.”
AI-based algorithms can be extremely sophisticated, and can al- low video analytics to operate in a way that was not possible in the past. Objects can be recognized and classified, along with their attri- butes, and stored as metadata (data about data) alongside the video, or in a separate system to aggregate the metadata to visualize trends and anomalies.
The algorithms can “learn,” so we are able to teach them a vast range of things that might be of value (if you have sufficient example images from which to learn). In the near-term for camera-based AI this will begin with identifying objects like general vehicle types, li- cense plates, people, and faces. It will continue to evolve soon there- after to also pick up descriptions of an object such as the color of a garment or if a person is carrying something or wearing a cap or glasses. And from there on to extracting object behaviors.
18
MARCH 2021 | SECURITY TODAY
COVER STORY