Organizations are deploying surveillance solutions in environments that were once closed off to delicate computer technology. Rugged NVR computers significantly expand the capabilities of edge surveillance deployments. They bring high-performance processing of stationary CPUs closer to the cameras to reduce network strain and signal latency. The proximity equals performance through real-time actionable insights and automatic responses informed by captured video.
Rugged NVR computers with powerful GPUs supply the processing muscle for a growing number of video analytics applications. Video analytics enhance the safety and security of surveilled areas in real-time by identifying dangers and discrepancies with IoT sensor triggers. Rugged NVR computers running video analytics can transform edge surveillance deployments into process optimizing, revenue generating systems. The following are five versatile video analytics deployments that offer useful insights from surveilled environments across several industries.
1. Facial Recognition
High performance GPUs use sequential and parallel compute for biometric identification deployments. The leading application for biometric processing — facial recognition — processes various features, textures and shapes of a subject’s face for immediate cross-referencing with a secure database. AI and machine learning improve facial recognition speed and accuracy, far exceeding the capabilities of human monitoring by validating with a deep neural network for machine intelligence.
A rugged NVR computer performing facial recognition processes at an unmanned airport security touchpoint can provide a finer screening with a highly accurate person-of-interest identification rate. Over time, system capabilities improve with the ability to identify subjects at greater distances with more data and information.
Thermal cameras used for facial recognition can form additional layers of security at identification touchpoints. Long-wave thermal infrared (LWIR) cameras capture temperature data from subjects, providing accurate images of head and face shapes that may visually be obscured by glasses, facial hair or deceptive prosthetics, even in low-light environments. Rugged NVR computers processing a combination of images from standard visual and LWIR cameras offer stronger security than either input alone.
Facial recognition can also be applied to customer service in many applications. Management may be alerted when priority customers or exclusive club members enter the premises, allowing them to deliver personal attention to the patrons. Acute facial recognition identifying signs of agitation in customer waiting areas may alert management to redirect resources to alleviate congestion and reduce wait times. Retail establishments, casinos and large venues can all leverage facial recognition to strengthen customer loyalty through proactive attention.
2. Medical Intervention
Rugged NVR computers applying much of the same facial recognition technology used at security touchpoints can also leverage the applications toward medical intervention. Visual and LWIR cameras can supply the rugged NVR computer with image data to be analyzed for subtle physical signs of illness or elevated temperatures.
Rugged NVRs can concurrently process the same image data from camera systems positioned for security monitoring to passively screen subjects for illness symptoms. By identifying subjects for additional health screening prior to accessing densely populated areas like public centers, transportation hubs or large venues, rugged NVR computers provide a powerful tool in the fight to contain widespread illnesses.
3. License Plate Recognition
High-resolution cameras deployed at facility parking locations and along roadways can instantly scan volumes of license plates to locate and track specific vehicles or maintain a record of vehicles entering or leaving an area. Rugged NVR computers running optical character reading algorithms accurately capture and process license plate digits automatically when the predetermined parameter is triggered, like a vehicle entering the camera field of view or a specific point within. The same rugged NVR computers are also integrated in first responder vehicles like police cars to process and store live camera footage. Police cars use these type of embedded computers as a control hub for cameras around the vehicles to wireless connected body cameras located on the police officer.
Business facilities can deploy license plate recognition to automatically grant gate access to vehicles displaying a plate registered in an approved visitor database. In ungated areas, the system can passively maintain a record of all vehicles that enter or depart a facility, noting dates and times, for accountability purposes.
K-12 educational facilities can use license plate recognition analytics to better safeguard students by logging vehicles entering the grounds and which students enter the vehicles, securing investigative leads in the event of abduction or missing person situation. For higher education, license plate recognition is a sound tool for parking enforcement when integrated with lot camera systems.
In addition to improving security, smart cities deploying license plate recognition camera systems can factor traffic patterns based on vehicle volume and repeat commutes. Traffic signals can be optimized on license plate data to correspond with anticipated traffic increases based on regularity and concurrence of commutes.
4. Behavioral Prediction
The rugged NVR computer GPUs can power behavioral prediction analytics that allow for quicker response times. The ability to anticipate how other parties may act can be directed toward avoiding unfortunate or harmful scenarios, optimizing processes and personalizing customer experience.
In-vehicle rugged NVR computers can alert truckers to signs that nearby vehicles may cut in front of them, brake suddenly or create any other dangerous situation requiring immediate attention. Machine learning refines traffic inference analysis, improving system response times. Behavioral predictive algorithms identify signs of distracted or impaired drivers on the road sooner, gradually allowing truck drivers more warning.
In-cab cameras capturing driver behavior and rugged NVR computers implementing a robust mix of analytics can even collate vehicle telematics like speed, positioning and braking with images of the person behind the wheel to note signs of dangerous driving in the truck drivers themselves. Algorithms used to identify facial features can easily detect signs of drowsy or sleeping drivers, prompting an alert to rouse the driver and direct them to take corrective action.
Behavioral predictive analytics have myriad security applications. Public transport, smart cities and venues can deploy rugged NVR computers in any surveillance environment to single out persons displaying very subtle signs of disruptive behavior. Airports applying behavioral predictive surveillance can identify traveler aggression that might be quickly de-escalated through prompt security intervention.
5. Event Detection
Rugged NVR computers deployed with event detection analytics use many of the same algorithms for the applications listed above. Event detection can alert smart city operators to numerous conditions that may lead to hazards for citizens, like unattended packages in public spaces, fallen items in roadways and weather-related complications.
The same preventative insights smart cities glean from event detection can be applied in virtually any environment where specific programmed visual triggers alert operators to changes from normalcy. A person or object falling onto tracks at subway and rail terminals may alert traffic coordinators and response teams to redirect or suspend travel until the situation has been resolved by responding security teams.
Schools, public buildings, large venues and shopping centers can all benefit greatly leveraging object-motion detection analytics to immediately identify firearms in the setting. Hypersensitive gun detection algorithms instantly alert security or law enforcement personnel upon the event of a weapon being drawn. This event detection analytic transforms a chain notification process relying on persons under stress to actively initiate a security process to a passive, real-time alert when a weapon is presented. By shaving valuable minutes or seconds off police response time, gun detection technology enables quicker intervention that can prevent disasters.
Rugged NVR Computers for Intelligent Edge Surveillance
Rugged NVR Computers for Intelligent Edge SurveillancePremio’s rugged NVR computers contain NVIDIA GPUs that rapidly process video data for a host of analytic applications to generate nearly prescient insight to IoT surveillance solutions. Video analytics deployed on these type of edge computers transform traditional recording and logging systems into powerful tools for proactive responses, streamlining processes and delivering more customized personal experience.
These specialized edge computers are built fanless; cableless; and with high tolerance for temperature extremes, moisture and physical abuse that allow them to be deployed closer to camera systems in challenging environments. Built to endure, the devices operate continuously in harsh conditions ranging from freezing outdoor locations to searing steel mills to vibrating and jolting rolling stock trains.
The ACO series of in-vehicle computers effortlessly record and process video in remote and moving surveillance deployments for a dedicated edge surveillance system. Mobile connectivity consolidates workloads by allowing in-vehicle rugged NVR computers to push video to security operation centers while receiving road or rail conditions information and detour instructions from the dispatch center.
The ACO 6000 series serve as electrical supply for low-power IP cameras through 16 PoE+ connection ports. With PoE+, electricity is delivered over the same cable transmitting video and signals between the camera and NVR. This flexibility allows strategic camera placement dictated by coverage needs, not power availability. Other PoE+ peripherals like access control hardware and alarms can be integrated into a holistic edge security system.