Industry
Smart cities and traffic
Highway and junction enforcement, with plate-linked evidence for traffic police.
Number-plate recognition
ANPR and Vehicle Intelligence
Automatic number-plate recognition (ANPR) tuned for Indian plates. It reads plates by day and night, detects helmet, triple-riding and wrong-way violations, and checks vehicles against a watchlist.
Vehicle records from camera video
This is the Videolytical product for roads, gates and car parks. It combines number-plate recognition with vehicle analytics.
ANPR and Vehicle Intelligence finds the number plate in a camera's video, reads its characters and logs the result with the time and the gate.
It also analyses the vehicle itself: direction of travel, parking and vehicle class. At high-security gates it supports under-vehicle inspection.
It has three main uses.
- Traffic enforcement: a violation is recorded with the plate as evidence for an electronic traffic penalty notice (e-challan).
- Gate and parking control: a vehicle entering or leaving is logged with its plate, the time and the gate.
- Watchlist alerts: a plate on a list of stolen or flagged vehicles is matched when a camera reads it.
Helmet violation detected, city highway.
What it detects and records
Capabilities are chosen camera by camera. A road, a gate and a car park each call for a different set.
- Number-plate recognition
- Finds the plate in live video and reads its characters, by day and night. Recognition is tuned for Indian plates.
- Watchlist matching
- Checks a read against a watchlist, also called a hotlist, of stolen or flagged vehicles. The log shows a watchlist status for each read.
- Traffic violation detection
- Detects riders without a helmet, triple riding and vehicles travelling the wrong way. Triple riding means three people on one two-wheeler.
- Plate-linked evidence
- Links a detected violation to the plate of the vehicle. The result is an evidence record for an e-challan workflow, which issues the penalty notice.
- Direction, parking and classification
- Reports which way a vehicle is moving, how long it stays parked and what class of vehicle it is. Typical uses are one-way lanes, parking areas and pick-up zones.
- Under-vehicle inspection
- Supports inspection of the underside of vehicles at high-security gates, where a vehicle is checked before it is let in.
What the operator sees
The ANPR dashboard, shown at the top of this page, follows a read from the camera stream to a row in the log.
-
Streams arrive from each gate
Each entry and exit gate has its own live stream on the dashboard, tagged in or out.
-
The plate is read
The live plate view shows a close-up of the current plate. The plate queue shows a card for each read: a crop of the plate, a wider snapshot and the text that was read.
-
The read is logged
The OCR (optical character recognition) log adds a row for the read: time, gate, plate text, vehicle colour, vehicle brand and watchlist status.
-
Watchlist matches raise an alert
Each read is checked against the watchlist of stolen or flagged vehicles. A match can be sent on as a real-time alert.
The same application has sections for users, the watchlist, cameras, reports, statistics and configuration.
What plate recognition needs from a camera
These points hold for number-plate recognition in general. Raise them at the site survey and camera audit, the first step of a rollout.
- Angle: the camera should face the plate as squarely as the site allows, both sideways and from above.
- Plate size: the plate should fill enough of the picture for its characters to be read.
- Shutter speed: a fast shutter keeps the characters sharp on a moving vehicle.
- Night capture: infrared lighting and headlight glare affect how plates read.
- Lane coverage: for enforcement, one camera for each lane is typically more dependable than one wide view.
Where the software runs
ANPR and Vehicle Intelligence runs on the Videolytical platform: at the edge, on-premise or in the cloud. The options can be mixed in one estate.
| Deployment | Where it runs | What it offers | Typically suits |
|---|---|---|---|
| Edge | On the camera or an edge box | Keeps working without connectivity | Roadside and remote gate cameras with a limited network link |
| On-premise | GPU servers on the customer's network | Video stays inside the customer's network | Sites that must keep video on site, including air-gapped sites |
| Cloud | Cloud instances | Central dashboards, updates and storage | Estates with several sites, managed from one place |
How it connects to other systems
The product uses the integration points of the Videolytical platform, so its events can reach systems your staff already use.
- Cameras and recorders
- Takes video from existing IP cameras and network video recorders over ONVIF and RTSP.
- Alert routing
- Sends alerts to the video management software (VMS) console, by SMS, by email or to a webhook. SMS and email alerts follow escalation rules and schedules.
- REST API and webhooks
- Pushes events and metadata to dashboards, and to security information and event management (SIEM) and physical security information management (PSIM) systems.
- Enforcement dashboards
- Passes violation evidence, with the plate read attached, to traffic enforcement dashboards and e-challan workflows.
- Existing consoles
- Shows Videolytical analytics inside an existing VMS or command-centre console.
- Users and access
- Supports sign-on through Active Directory (AD) or LDAP. Access is role-based, with permissions set for each camera.
Where it is used
Typical deployments are on city roads and at transport hubs. These pages give the detail.
Industry
Airports and transport
Number-plate recognition for parking and pick-up zones at airports, railway stations and bus depots.
Solution
Traffic enforcement
How violation detection and number-plate recognition work together on the road.
Case study
City highway helmet enforcement
Rider helmet detection with ANPR evidence capture on highway cameras, by day and night.
Datasheet
ANPR and Vehicle Intelligence
Automatic number-plate recognition and vehicle analytics software for roads, gates and car parks. Tuned for Indian plates, by day and night.
| Product | Automatic number-plate recognition and vehicle analytics software |
|---|---|
| Plate recognition | Number plates read by day and night, tuned for Indian plates |
| Violation detection | No helmet, triple riding, wrong-way driving |
| Watchlist | Matching against a watchlist, or hotlist, of stolen or flagged vehicles |
| Vehicle analytics | Vehicle direction, parking and classification |
| Gate security | Under-vehicle inspection at high-security gates |
| Evidence | Plate-linked evidence records for e-challan workflows, which issue electronic traffic penalty notices |
| Operator dashboard | Live plate view, plate queue, live streams for entry and exit gates, and an optical character recognition (OCR) log |
| Log fields | Time, gate, plate text, vehicle colour, vehicle brand, watchlist status |
| Application sections | Users, Watchlist, Cameras, Reports, Statistics, Configure |
| Camera support | Existing IP cameras and network video recorders over ONVIF and RTSP |
| Alerts and interfaces | Alerts to the video management software (VMS) console, by SMS, by email or to a webhook. REST API and webhooks for events and metadata |
| Deployment | Edge, on-premise or cloud. The options can be mixed in one estate |
| Sizing | Network and GPU capacity are determined during the site survey |
Videolytical Systems Pvt. Ltd., 277, C Block, Sector 63, Noida, Uttar Pradesh, India. Sales and demos: sales@videolytical.com, +91 97160 05441.
Questions buyers ask
Does it work with the cameras we already have?
It takes video from existing IP cameras and network video recorders over ONVIF and RTSP. Whether a particular camera gives readable plates depends on its position, angle and lighting. The site survey and camera audit, the first step of a rollout, looks at the cameras you have.
Does it read plates at night?
Yes. Number-plate recognition works by day and night. Night results depend on the camera and its lighting, so include a night clip from your own site when you evaluate.
Does it need an internet connection?
No. At the edge or on servers at your site, the software keeps working without outside connectivity. With an on-premise deployment, video stays on your network. Cloud is an option for estates with several sites.
How is e-challan evidence produced?
Each detected violation is linked to the plate read and becomes an evidence record for an e-challan workflow. Our smart city case study describes this for rider helmet detection on highway cameras.
How much hardware does it need?
That depends on the number of camera streams and on where the analytics run. In a rollout, network and GPU sizing follows the site survey and camera audit.
How can we evaluate it on our own site?
Ask for a live demo on your own footage, or send a sample clip from a road or gate camera. The usual next step is a pilot on a small set of cameras before scale-out.
See ANPR on your own footage
Arrange a live demo, send a sample clip from a road or gate camera, or ask for an architecture and sizing discussion.