Pixuate
Traffic & Road Safety Analytics

Vehicle Count Classification

Automatic Vehicle Count and Classification (AVCC) product that counts and classifies every vehicle passing through a camera view. Provides directional flow data by vehicle class across roads, toll plazas, and city corridors — used for traffic management, toll operations, and urban planning.

Key Features

Discover the powerful capabilities that make Vehicle Count Classification stand out

Counts vehicles by class — car, motorcycle, bus, truck, LCV, and more

Directional flow — counts per lane and direction of travel

Time-stamped aggregation by hour, shift, peak period, and day

PCU (Passenger Car Unit) conversion for planning outputs

Toll classification support — matches vehicle class to toll bracket

Real-time dashboard and historical trend analysis

Exportable count data for traffic models and reporting

Edge deployment — continuous operation without cloud dependency

Technical Specifications

Detailed technical information about Vehicle Count Classification

Camera:2MP+ IP CCTV (RTSP), overhead or side-mounted
Deployment:Edge device or on-premises server
Output:Vehicle counts by class, lane, direction, and time window; PCU summary

Prerequisites & Limitations

What Vehicle Count Classification needs to run reliably, and the operating limits to plan for.

Prerequisites

  • Camera: 2MP+ IP/RTSP, overhead or side-mounted, covering all lanes to be counted. Capture speed is camera-dependent — up to ~80 km/h with a standard IP camera; a global-shutter camera for higher speeds
  • Adequate illumination — pair with an IR illuminator for night counting
  • Stable bandwidth between camera and edge/server — roughly 4–8 Mbps per 2MP H.264/H.265 stream
  • Mounting for full lane coverage and clear separation between vehicles
  • Stable power and network at the roadside or gantry

Operating Limits

Factor Edge Server
Incident / event capture rate≤ ~5 events per second≤ ~20 events per second
Concurrent camera streams per node1–4 streams (Jetson TX2 / Orin)8–16+ streams
Minimum object/plate size in frame≥ ~20 px≥ ~20 px
Event-to-alert latency< ~1 s on-deviceDepends on network round-trip

Indicative figures — actual limits depend on hardware, scene, and configuration.

Related Knowledge Base

Learn about the hardware and infrastructure components used in a Vehicle Count Classification deployment.

Product demo

Ready to Get Started with Vehicle Count Classification?

Contact our team to learn more about this product and how it can benefit your organization

Ready for deployment

Ready to Transform Your Operations?

Join hundreds of forward-thinking companies using Pixuate to unlock the power of video data. Subscribe for updates or get in touch for a demo.

Edge AI analyticsEnterprise supportDeployment guidance

Stay Updated

We respect your privacy. Unsubscribe at any time.

Looking for a custom solution?