Remember the old adage “The whole is greater than the sum of its parts?” Nowhere is that truism more evident than when you add network video to the current generation of Internet of Things (IoT) solutions. Whether we’re talking about industrial IoT applications, “Smart – X” (city, building, parking etc.) or retail operations, integrating network video into the solution provides value far beyond simple situational awareness.
Optimising sophisticated video technology
When video systems first moved from analogue to digital and then became part of the IoT world, they were primarily used to provide visual validation of sensor-detected events. For instance, if an industrial controller sensed an environmental issue such as a temperature exceeding set threshold maximum limits, the sensor would trigger the management software to notify the operator that this event had occurred. The operator could then pull up the video feed of the closest camera and observe the area remotely. While this application is simple, it shows how video enhances sensor management.
As edge devices, such as sensors and network video become more intelligent, the interactions between systems are growing in sophistication and generating even greater value than each system could provide on its own.
To appreciate how these smart applications are being used to improve overall efficiencies and profitability, let’s delve into three areas where they’re being deployed: intelligent buildings, smart cities, and smart retailing.
| By overlaying intelligent operational sensors with intelligent video, it’s now possible to automate lighting levels based on motion detection |
Video-based operational analytics
Applying intelligent monitoring to environmental equipment (HVAC) makes it easy for building owners and property managers to determine existing operating costs based on current equipment performance. They can then compare that amount to the cost of upgrades and potential cost savings over time.
Lighting is another significant operating cost within building management. By overlaying intelligent operational sensors with intelligent video (light sensors), it’s now possible to automate lighting levels based on motion detection. Lights can automatically turn on or off, brighten or dimmed, to eliminate wasteful energy consumption. With the addition of occupancy analytics via intelligent video, property managers can determine what caused the motion and learn other operational details such as occupancy counts. Did someone walk through and area causing lighting to turn on or up? Did they dwell in this area? These specifics can help managers efficiently optimise lighting controls and reduce the overall operating cost of the property.
Businesses are also using smart applications to optimise allocation of desk space and conference areas. For instance, intelligent video can determine conference room occupancy (in use, number of people in room, free space even though showing booked) far better than stand-alone motion sensors. When tied to automated room assignment systems, the additional statistics provided by video analytics might suggest room changes based on room size and number of attendees through back-office applications such as Microsoft Outlook.
These examples are just a few of a growing list of available video-based operational analytics currently on the market.
Video analytics in smart cities
Initial forays into smart city technologies such as smart lighting, smart grid, smart parking and so on relied on standalone sensor technologies. Their capabilities were good but limited. Smart Lighting for instance would use basic light detectors to turn street lighting. Smart Parking and traffic systems would use weight sensors to trigger vehicle counts, traffic signal changes or determine if a parking space was in use and paid for. Augmenting these applications with intelligent video and analytics, however, opens up a whole new world of additional details. In Smart Lighting, the video sensor can now trigger a change in lighting based on rules such as vehicular and pedestrian events. Video analytics can yield additional metadata such as vehicle type (commercial versus public use). Smart Parking becomes much more effective when you can begin to provide vehicle detail such as vehicle type or other information based on licence plate recognition. These additional details can help parking lots operate more efficiently and offer value-added services like space reservation and open space location notifications.
| Augmenting smart city applications with intelligent video and analytics opens up a whole new world of additional details |
Smart Grid offers some less obvious but equally valuable system augmentation capabilities. We often associate Smart Grid with simple automated meter reading but these systems also traverse critical power infrastructure. Solution providers in this arena are now offering heightened asset and perimeter protection via integration of network-based radar detection with video and audio analytics. This strategic mix of technologies can be used to minimise false detection alarms, turn on/off or change lighting levels and point cameras to areas of interest for extremely effective and cost-effective perimeter security.
Network video for retail intelligence
Retailing was one of the earliest adopters of smart device integration with network video and video analytics to support loss prevention and customer safety. They’ve been using video to analyse customer traffic and behaviour in order to improve product placement, increase product sales, as well as cross-sell related items. Adding programmable “Digital Signage” to the mix created new opportunities to display targeted messages based on viewer demographics about additional products and services of potential interest.
Integrating network video with point-of-sale terminals to reconcile cash register receipts, adding heat mapping analytics to study customer foot traffic patterns, measuring check out wait times to increase employee productivity and efficiency as well as improve the customer experience are just some of the ways retailers have applied the principles of IoT to their advantage. Overlay intelligent building controls and you can see the exponential power of integrating intelligent video with other IoT devices and systems.
| Retailing was one of the earliest adopters of smart device integration with network video and video analytics to support loss prevention and customer safety |
Minimising metadata overload
Smart application integration produces an enormous amount of metadata. Collecting, transporting and synthesising this data into meaningful business intelligence can be daunting. It requires disciplined use of resources from the network infrastructure transporting the data locally to the various cloud technologies (private cloud, hybrid cloud, public cloud) storing and disseminating it securely.
Generally smart sensor data is fairly light weight in terms of actual data transmitted. Adding video elements can significantly increase bit-rate (bandwidth and storage) requirements. This highlights the need for the video to be more intelligent and interactive with the intelligent sensor and edge device technologies so that resources can be used more efficiently. Smart applications let you do that. You can fine tune video rules and optimise transmission based on retention value. You can program the video to sensor triggers or events, transmitting lower frame rate and resolution video for less interesting video and increasing the video settings when higher quality video is more relevant and valuable based on these sensor triggers.
The back-end collectors of sensor metadata are becoming more mainstream and easier to operate. In many sectors, service providers are offering management of this sensor output “As a Service.”
As smart IoT technology continues to mature, the benefits of integration between network video systems and other network solutions will only get better. We’re already seeing greater efficiency in operations as well as higher quantifiable returns on investment through cost savings and more in-depth, usable business intelligence.