Cloud and Analytics: A Powerful Combination
By Aaron Saks
Sr. Technical Marketing and Training Manager
Hanwha Vision
Two of the most rapidly growing areas in security and surveillance are the cloud and intelligent analytics.
Cameras of all types are now being equipped with more advanced analytics, and the role of on-board analytics will continue to expand as users develop new needs for more accurate object tracking or to generate data-driven business insights.
With the cloud, there are still some concerns and uncertainty regarding security, reliability, and access. But as more end users grow familiar with using the cloud for storage and sharing, the trend of cloud adoption will continue to increase, especially as bandwidth increases and costs come down.
Combing Cloud with Analytics
One more term we need to discuss is the “edge.” The edge may refer to points of presence (PoPs) close to network endpoints, or connectivity that brings recording and storage closer to the data source or device. Security at the edge is much more effective, and the fact it's decentralized gives organizations more options for managing their own unique security requirements. Perfect for standalone or remote systems, edge-based recording only uses a network for playback and review, conserving bandwidth and allowing more stable and reliable recording of high-quality, high-resolution content through on-board flash media. The addition of intelligent analytics further enhances the long-term value of stored content.
The use of Edge AI, especially with analytics based on deep learning algorithms, can be a key element in a range of smart surveillance applications. These include object detection and classification, and collection of attributes in the form of metadata—all while reducing latency and system bandwidth burdens and enabling real-time data gathering and situational monitoring.
The security and surveillance industry has traditionally experienced two major pain points. One is making devices work together seamlessly from an IT perspective. The second has always been increasing the effectiveness of analytics to eliminate, or at least significantly reduce, false positive alarms that waste security teams’ time and lowers efficiency.
Now, with AI and the addition of Deep Learning algorithms, analytics can truly do what the user needs instead of giving them hundreds of notifications triggered by trees blowing in the wind instead of the objects or people they really need to track.
For example, people typically want the convenience of easy notifications pushed to them. That's where the cloud is ideal. Some systems require users to be logged into the cloud to be able to get push notifications. With AI-based analytics on the edge, now we can push those to the “other” edge, to the user side by using the cloud as essentially a conduit.
That type of system enables easy connectivity without the hassle of designing that connectivity or going through a VPN. The benefit is being able to take cloud-enabled data and then connect it easily to its compatible client, like a mobile app, so both work together. That's one of the biggest benefits to the cloud: You can have a cloud video system up and running quickly. You can take your intelligence—your analytics—with you on the road without having to be tied to a specific place to view them.
Cloud-based services also give users added flexibility and scalability to customize and manage their services according to individual needs. For example, some cloud platforms, including Hanwha’s Wisenet Sky, offer a subscription model allowing users to choose the level of services they need at certain times of the year or by project.
If you’re only recording and storing for 30 days now, but you know that in a few months or seasonally you’ll need 60 days of storage, increasing cloud storage is a matter of logging in and clicking a few buttons. Users don't need to go on site and change out hard drives, since the cloud is elastic and scales to needs and resources.
This is becoming a more common way of working, especially as customers get more comfortable with the cloud and less concerned about security and reliability.
There are many applications that can dictate when a hybrid approach makes sense. For example, in many markets such as cannabis or banking, there are regulations stating you must have both on-site and off-site back-up storage. Hybrid allows full on-site video recording capability combined with a full cloud user interface with all the necessary features and functionality, and without tying up network bandwidth. If you don't need to send all your data to the cloud all the time, then hybrid gives you the best of both worlds. You can mix and match or go either way.
Storage to the cloud can even be scheduled at optimal times, maybe in the evening when fewer people are in the office and not all fighting for bandwidth. Choosing your cloud path is easier now than ever. There are dozens of cloud storage providers available that are credible, reliable and offer scalable computing.
Certain markets are still cautious when it comes to the cloud, especially if the market has strict regulations about how surveillance data is stored, for how long and where. Also, certain markets have varying needs for analytics, for example, tracking different objects or data.
Schools using cameras with built-in analytics can perform people-counting to accurately track the volume of people entering or exiting a building, which helps building administration monitor their population on any given day against their attendance records. Digital imaging surveillance technology combining AI with on-board audio and video analytics can help school administrators get a better handle on access control and monitoring of hallways, classrooms, and exterior parking lots.
Healthcare professionals need to monitor patients to determine if they are secure in bed or at risk of falling out. Also, healthcare facilities, or any building with sensitive areas, may keep building “watch lists” of people whose previous histories prohibit them from being on the grounds. Those individuals can be identified by their license plates while they’re still in the parking garage and security teams can intercept them before they enter the facility. If they do get past the front doors, their movements can be tracked throughout the building.
A market like cannabis can be tricky because there are so many different rules that vary by state. Some states require offsite backup of security archives, but the “off-site” requirement can be satisfied by storing data next door in a separate building not physically attached. In other cases, offsite may mean the only option is to use the cloud.
When used separately, the cloud and analytics are effective tools, but when combined, their shared benefits contribute to a total integrated security and surveillance program.
Aaron Saks is the Sr. Technical Marketing and Training Manager with Hanwha Vision.