Number Plate Reader

Overview

Number Plate Reader detects license plates in an input frame, reads the text from each plate, and returns decoded text, detection polygons, and confidence scores.

Use this node in vehicle access control, parking management, tolling, or any flow that needs to extract plate text from camera frames.

Input

Input Image

image required

The image frame to analyze. Connect this to a camera or upstream image output.

Detection Model

string required

Balances processing speed against recognition accuracy.

Values:

• SPEED (default) — optimized for real-time frame rates; suitable for high-speed lanes or high-frame-rate cameras.
• ACCURACY — optimized for maximum character recognition accuracy; suited for lower-frame-rate or still-image scenarios.

Character Type

string required

Expected character set on the plates, used to constrain the recognition model and reduce errors.

Values:

• NUM — digits only (0–9).
• NUM_LATIN (default) — digits and Latin letters (A–Z).
• NUM_THAI — digits and Thai characters.

Choose the character type that matches the plates in your region.

Draw Text

boolean required

When enabled, draws the decoded plate text directly on the plate region in the overlay.

Overlay Results

boolean required

Whether to draw detection polygons and labels on the output frame. See Overlay Results.

Output

Overlay Image

image

Output frame from the node. If overlays are enabled, plate detection polygons, decoded text, and confidence scores are drawn on this frame.

Detected Count

integer

Number of license plates detected in the current frame.

Detected Plates

array

Array of plate detection objects. Each object contains:

• quadrilateral array: Four corner points of the detected plate polygon in image pixel coordinates.
• label string: Decoded plate text.
• confidence number: Recognition confidence score.