[ 03 ] Smart city
Edge AI for smart cities and traffic
City and roadside cameras that do more than record: they monitor traffic, detect violations, warn of floods and spot infrastructure problems, with processing at the edge so video does not have to be streamed to a central cloud.

Who it’s for · Municipalities and city districts, traffic police and transport authorities, utilities, and smart-city system integrators and telecom operators.
[ 01 ]Why Nabrio in smart city
Built for how your sites actually work.
- 01
Processed at the edge
Nara runs on NVIDIA Jetson devices or on-site servers, which reduces bandwidth and keeps video local.
- 02
One platform for the city
Traffic, public safety and environment monitoring use the same platform and the same building blocks.
- 03
Tracks across devices
Multi-device sync shares data between Nara devices, so people and vehicles can be followed across locations on a site.
- 04
Feeds the command centre
Results and evidence go to command-centre systems over MQTT or HTTP.
[ 03 ]Featured use cases
What smart city teams build on Nara.

cameranara seesTraffic monitoring
Cities need live traffic data and early warning of congestion and incidents without streaming every camera to a central cloud.

cameranara seesTraffic violation detection
Enforcement teams cannot watch every junction, and violations need reliable evidence.

cameranara seesLicence plate recognition
Vehicles entering and leaving a site are logged by hand at the gate, and parking and enforcement need a reliable plate record.

cameranara seesPeople counting
Simple door counters double-count people who return or use several entrances.
[ 04 ]Every use case
The full list.
Everything in this sector that Nara can do, grouped by area. Use cases with their own page link through.
Traffic monitoring
- Traffic flow and congestionVehicle detection + tracking → counts per lane and direction, density, queue length, congestion alerts
- Vehicle classification countsCars, motorbikes, buses, trucks per hour
- Traffic accident detectionDetect stopped vehicles in live lanes, collisions, and crowds forming → alert the command centre
- Licence plate recognitionPlate detection + OCR for enforcement, access, and investigations
Traffic violations
- Illegal lane changeTrack crosses a solid-line boundary
- Bus-lane violationA non-bus vehicle in the bus-lane zone (during restricted hours)
- No-parking zoneA vehicle stationary in the zone longer than a set time
- No-stopping zoneA vehicle stops in the zone for longer than a set time
- Parking violation (double parking, wrong bay)Zone + dwell time + vehicle type
- Wrong-way drivingTrack direction against the allowed direction
- Motorbike or car on the pavementA vehicle track inside the pavement zone
- Crossing the pedestrian crossing on redVehicle in the crossing zone while the pedestrian signal is green (needs signal state)
- Red-light runningStop-line crossing while the signal is red (signal state from the controller or camera)
- Over-speed detectionTracked distance over time with a calibrated camera
- Evidence captureEvery violation: plate read, images, timestamp, location to the database or back office
Public spaces and safety
- Hawker (street vendor) violationDetect stalls and carts in restricted zones and hours
- Crowd and overcrowdingCrowd-counting model → density alerts for events and stations
- Fights and violenceAction recognition → alert patrols
- Falls and people lying downPose and fall models → emergency response
- Illegal dumpingDetect dumping activity or new waste piles in watched areas
- Smart pedestrian crossingsDetect waiting pedestrians → adjust the signal timing
Environment and infrastructure
- Flooding detectionDetect water on roads and underpasses → warnings and road-closure alerts
- River and canal water levelRead a staff gauge or water line from the camera → level trend, threshold alarm; fuse with level sensors over Modbus or MQTT
- Weather monitoringClassify sky, rain, and visibility conditions; combine with weather sensors
- Street-light and blackout monitoringBrightness per zone at night → detect lamps that are out or area blackouts
- Pothole and road-damage detectionCameras on vehicles or fixed cameras → detect and geo-log damage for repair crews
- Bridge and utility inspectionDrone or video feeds → detect cracks and rust; log for maintenance
[ 05 ]Integrations
Connects to your systems.
Roadside and junction cameras, NVIDIA Jetson edge devices or on-site servers, and command-centre or back-office systems over MQTT or HTTP.
All integrationsFAQ
Smart city: questions
Does video have to be streamed to a central cloud?
No. Nara processes video at the edge and sends only results and evidence to the command centre.
Which violations can it detect?
Lane, bus-lane, no-parking, no-stopping, parking, wrong-way and pavement violations use built-in tracking, lines, zones and timers. Red-light, pedestrian-crossing and speed cases also need the signal state or a calibrated camera.
Plan a proof of concept for your smart city site.
Download Nara and build your first flow in minutes, or tell us what you need to see and do, and we’ll show you how it runs on your own hardware.