Higher-performance IoT devices, advanced deep learning models, the growth of artificial intelligence at the edge—as smart city technologies become more sophisticated, public safety IoT is entering a new era of heightened awareness; real-time analysis; and autonomous, risk-reducing systems.
The Need for Smarter Public Safety Technology
Public safety agencies—law enforcement, fire, EMS, and emergency management—have always been early adopters of the latest technologies, from the earliest telegraphic fire alarms to radios and dash cams. Today, expanding networks of smart devices and the rise of artificial intelligence are transforming public safety technology from an essential toolset into an autonomous partner that can act to help keep the public safe.
Smart Public Safety Technologies Are Already at Work
- Smart streetlights with cameras, microphones, and sensors are using computer vision to gather intelligence about traffic, accidents, and crime. In a crisis, they can call for help and direct people to safety.
- Smart traffic signals are improving urban mobility by analyzing traffic conditions, changing their timing, and easing congestion. During emergencies, they can automatically give first responders the right of way.
- Smart intersections are capturing and analyzing traffic patterns, detecting risks, and warning against imminent accidents, reducing crashes and injuries.
- Smart emergency vehicles are capturing and sharing video, audio, and vehicle telemetry, giving dispatchers a real-time view of the field and invaluable data for training and planning.
- Smart buildings are monitoring access, reviewing video feeds, and running environmental systems. In an emergency, they can share camera feeds and building data with public safety officials.
According to Juniper Research, smart cities may see up to a 15 percent improvement in emergency response times and a 10 percent reduction in violent crime.1
Smart Cities and Intelligent Emergency Management
Smart cities integrate these public safety IoT devices using a common framework and shared data pools. This gives every agency a single pane of glass and shared view of situations as they unfold. It also opens the doors to APIs that can connect public safety technology stacks to citizen data sources like social media, news feeds, even smart building systems.
Benefits to Government Transformation
These intelligent, data-driven public safety technologies create a blueprint for transforming how city governments share information, make decisions, and serve their constituents.
- Open frameworks and shared data lakes help break down silos between departments.
- Cities can include private data from hospitals, businesses, and smartphones from citizens who opt in.
- Sharing information systematically keeps agencies, departments, and citizens on the same page.
- This objective data plus AI-assisted analysis give cities and their constituents a more precise understanding in the moment and over the long haul.
- Policy makers, businesses, and individuals have better information for making decisions.
- Cities can see how policies impact the community more quickly and accurately.
- The community gains a powerful, intelligent public/private network that can be put to work in a crisis.
The sheriff’s office in Wake County, North Carolina, used this strategy to extend digital services beyond their department. Their system serves as the central data hub for all the county’s public safety‒related data, which other agencies can access and use as needed. Sharing criminal justice and civil process data through this secure platform speeds communication and helps other government agencies work more efficiently. Moving to a single platform means the county no longer needs to manage solutions for each agency, which saves time, labor, and money.2
Smart Technologies and Incident Response
When an emergency occurs, smart public safety technologies become critical players in the response. Cameras on nearby smart streetlights and private buildings can provide more angles on the situation. Body cameras and smartphones turn each first responder into a live source of audio, video, and location data.
This incoming data gives incident commanders a real-time, street-level view and a more accurate understanding of the bigger picture. Commanders can also use smart city technology to support first responders in the field.
- Smart streetlights can flood an accident with light.
- Lights can change colors, blink, and broadcast messages.
- Smart traffic signals can clear the streets for first responders and detour traffic.
- Smart kiosks and digital signage can alert people to dangers and instruct them on what to do.
Disaster Preparedness, Response, and Community Resilience
Disasters—wildfires, hurricanes, earthquakes, terrorist attacks—unfold over longer time frames and impact far more people than day-to-day incidents. Smart public safety technologies can help city managers prepare for, respond to, and recover from major, traumatic events.
Day-to-day intelligence from public safety IoT and smart city devices gives policy makers the objective information they need to identify issues, improve public safety, and build a stronger, healthier community.
During a major crisis, smart devices can help emergency managers maintain an accurate view of the situation and make informed decisions. Artificial intelligence can analyze the flood of data and separate the most important signals from the noise. AI can even help triage response by prioritizing critical communications and pinpointing locations that need resources the most.
After a disaster, smart public safety technology can help identify which neighborhoods need water and supplies, which streets are unsafe to travel, and which buildings are structurally sound. Smart city technology can even help speed recovery by providing data the city needs to prioritize reconstruction, streamline insurance claims, and make strategic decisions about its future.
The impact of smart public safety technology on community resilience is not theoretical. Rio de Janeiro created a central command center with inputs from more than 30 city agencies. The shared information enables city agencies to map high-risk areas for floods and flood-related landslides and create an early warning and evacuation system.3
Public Safety Technologies in Law Enforcement
Objective information is critical for officers in the field and a vital record for the communities they serve. Public safety technology helps law enforcement and government provide more data, more transparently—a fundamental step in creating true community policing that brings officers, citizens, and decision-makers together.
AI is helping the city of New Orleans analyze evidence from more than 325 cameras. The system has helped police in 70 percent of cases by providing relevant video and information and saving the department more than 2,000 hours of manual labor.4
Next-Generation Public Safety Technology
The exponential growth of computing power at the edge is expanding capabilities and spreading artificial intelligence everywhere. Public safety IoT devices that can perceive, react, and intervene on their own are becoming an everyday reality.
Here are the key technologies that make it all possible:
Higher-Performance IoT in Public Safety
Making cameras, streetlights, and intersections smart requires high-performance computing and networking devices that can stand up to extreme conditions. Manufacturers are responding with embedded and industrial-rated systems on chip (SoCs), computers on modules (COMs), and purpose-built devices that can deliver artificial intelligence and autonomous computing to the most-demanding edge applications.
5G Edge Networks
Compared to 4G, 5G networks will deliver 10x less latency, 50x more speed, and 1000x more capacity.5 In an emergency situation, these extra seconds can help save lives. This dramatic increase in speed and bandwidth will put emergency services even closer to a real-time understanding of events on the ground so they can coordinate and respond even faster.
Deep Learning and Predictive Public Safety
In deep learning, powerful computers learn by recognizing patterns in data. Cities are using deep learning to analyze crime data and identify meaningful patterns.
The city of Manchester, New Hampshire, created a predictive policing system using crime statistics, weather patterns, and other information overlaid on a city map. The system predicts where crimes will occur within a 500-foot radius. It averages 60 percent accuracy and has helped law enforcement reduce total crime by 28 percent.6
Artificial Intelligence, Computer Vision, and Sensory Applications
Deep learning also creates the models that other devices use to understand what things mean. For example, data scientists have trained deep learning models to recognize the sound of breaking glass. Smart streetlights can use these models to listen for accidents and break-ins. If a smart streetlights hears glass breaking, it can turn itself red, blink, and call for law enforcement—all with no human intervention.
Artificial Intelligence gives smart city and public safety technology the ability to sense, analyze, and act.
- Smart cameras can see accidents and call for EMS.
- Microphones can identify gunshots, triangulate the shooter’s location, and relay it to first responders.
- Natural language processors can transcribe interviews and enter them into evidence management systems.
Gunshot detection technology helped catch a criminal who fired multiple shots at people in a Fresno, California, neighborhood in 2017. Officers were alerted almost immediately and captured the suspect before he left the area.7
Open Standards, Shared Data, and Heightened Security
Public safety data isn’t useful if first responders, commanders, and government leaders can’t access it. Open hardware standards, open data frameworks, and shared data pools are critical to the success of any smart public safety technology.
All that data comes at a price. Once they capture data, agencies become responsible for keeping it secure, especially sensitive criminal and medical data. As intelligent technologies spread, governments have to secure thousands of embedded devices and protect all the public and private data as it moves through the system.
Hardware-based data security measures in conjunction with hands-free, remote device management can help agencies thwart attackers and malware.
Intel® Reference Designs for Public Safety
Vehicle computers, intelligent cameras, and smart intersections are some of the technologies that make smart public safety technologies possible. Intel and our partners offer a wide range of hardware and software solutions for public safety, urban mobility, and smart cities. The Intel ecosystem provides prevalidated reference designs, reference implementations, and ready-to-run solutions for public safety.
Roadside Unit Reference Design
Our reference design for roadside edge computing can be attached to streetlights and other fixtures. This roadside unit is ideal for real-time video analytics and other performance-hungry tasks. Cities can deploy these units as part of a solution for smart streetlights, smart traffic lights, smart parking, or e-tolling stations. They deliver the processing capabilities needed to detect license plates, spot pedestrians, and monitor traffic congestion. These edge nodes can even provide public Wi-Fi coverage. Supported by Intel® Vision Products with integrated security features, the AAEON Atlas edge computing node is a ready-to-deploy solution based on this reference design.
Converged Edge Reference Architecture (CERA)
CERA is a platform approach for IoT and networking workload convergence. With this architecture, our partners can design solutions for roadside equipment to process sensor modalities and perform sensor fusion. This brings intelligence to the edge while hosting 5G network capabilities and microservices. Solutions built on this platform can be set up at intersections or on-premises for near-edge computing and data processing for multiple IoT devices. Solutions can be optimized using the Intel® Distribution of OpenVINO™ toolkit and OpenNESS toolkit. With 5G connectivity, CERA provides networking capabilities that allow IoT devices to communicate with each other at the edge or send data to the cloud.
|IoT and Embedded Intel® Processors||Intel® processors come in a range of performance and power profiles for intelligent cameras, sensors, and embedded computers for public safety.|
|Intel® Xeon® Scalable Processors||Intel® Xeon® Scalable processors deliver high performance for edge servers, ideal for performing real-time analytics and AI on smart road sensor data.|
|AI and Computer Vision|
|Intel® Movidius™ VPUs||Intel® Movidius™ VPUs enable computer vision for specific use cases, such as finding or “seeing” license plates and vehicles at smart intersections.|
|Intel-Supported 5G Networks||Intel-supported 5G networks will improve real-time traffic data at the edge while also advancing connectivity and transmission to and from wireless networks.|
|Intel® Edge Software Hub||Find software to accelerate the development of smart road infrastructure solutions, including referenced implementations for intelligent traffic management.|
|Intel® Distribution of OpenVINO™ Toolkit4||The Intel® Distribution of OpenVINO™ toolkit streamlines the development of vision applications on Intel® platforms, including VPUs and CPUs. This portfolio enables computer vision to locate pedestrians, cars, and street signs.|
|OpenNESS||OpenNESS open source software simplifies the complex orchestration and management of edge services across diverse network platforms and access technologies.|
|Intel® DevCloud for the Edge||Reduce time and costs in determining the right hardware for optimal AI application performance. Intel® DevCloud for the Edge provides instant performance feedback via a virtual AI prototyping tool.|
|Open Visual Cloud||Build media processing apps, including video on demand (VOD) and live streaming with SVT-AV1. This collection of open source stacks and pipelines includes optimized ingredients for encode, decode, inference, and render. It’s a reusable environment that eases testing, evaluation, and deployment.|