Sensorable ANPR camera in detail

Our ANPR camera is a speed and traffic monitoring device that needs minimal power and no data infrastructure.

Functional components

Sensorable smart camera

The current model combines a full color image sensor, CPU, GPU and 4G / WiFi / Bluetooth modems. Think of it as a tiny computer that does:

  • processing radar output
  • object detection
  • movement detection and trajectory analysis
  • video recording, compression and analysis
  • uploading data to the cloud database

Traffic radar

The camera relies on a connected radar to get timely speed readings on the passing traffic. There are multiple radar integration options:

  • an internal radar in the same enclosure as the camera
  • an external radar installed inside the speed sign
  • a 3-wire RS-232 or USB connection
  • multiple radar makes and models are supported
  • 2-way communication to re-configure the radar remotely

Power supply

Standard power supply sources:

  • an internal battery pack
  • a battery inside the speed sign (6v-24v)
  • street lighting or PoE via an adaptor

The peak power is drawn for a brief moment only while processing a frame with a vehicle. It averages out at approximately 0.02Wh per vehicle.

  • Peak power: 4W
  • Idle power: 0.12W

Vehicle processing steps

1. Read radar output

All speed readings and other info from the radar are captured and uploaded to the cloud DB.

2. Record the target vehicle

A vehicles moving over a set speed in a set direction is captured on video.

3. Read the number plate

The text from the number plate on the selected vehicle is extracted and checked for consistency across all frames.

4. Broadcast event details

A pre-configured message is sent to a connected LED board or another device via Bluetooth Low Energy, including the speed and number plate of the vehicle for immediate display.

The speed data, a small image of the vehicle and a separate image of the number plate are uploaded to the cloud, including the exact timestamp, GPS coordinates and device status.

Processing parameters

Default options

  • Processing speed: 3fps
  • Recording speed: up to 30fps
  • Still images up to 12 megapixel
  • Video up to 1920 x 1080 (Full HD)
  • Full color, ambient light
  • Vehicle selection: configurable path

Configurable options

  • Image and video resolution
  • Record stills, video or both
  • Upload stills, video, both or none
  • The ANPR step is optional
  • Broadcast to connected devices is optional
  • Over 50 other configuration options

The entire process of detecting vehicles, recording video and uploading data can be configured for privacy, cellular data allowance, power availability and other external factors. It can range from uploading everything immediately to uploading statistics and aggregate data on a schedule.

User interface

All the data uploaded to the cloud storage is verified on the fly, converted into different formats and sent to various other systems for further processing or viewing by users.


All the data collected from cameras is stored in a cloud DB. The amount of data to be collected depends on the application and your cellular data cost.

The UI presents the data as graphs, tables and raw logs. They can be viewed in a web browser or downloaded for further analysis.


Which media to be recorded and uploaded is configurable depending on the application and the cellular data cost. Photos and videos can go through automated post-processing after being uploaded. Users can view them in a web browser alongside the speed of the vehicle.

Actionable insights

The cloud system can email automated reports with a small number of high priority records for immediate action. E.g. top 5 fastest vehicles or a short list of consistently fast vehicles.

Power in simplicity

Our ANPR camera was designed for places like roadworks, school zones, suburban and rural roads with low to moderate traffic density.


The low hardware cost of the camera makes adding an ANPR feature to VMS boards and speed indicating signs affordable on a small and large scale.


Check out our GitHub repository for more detailed technical analysis of this product.