Plate Recognizer examples and utilities

This repository contains example clients, integrations, and operational utilities for Plate Recognizer Snapshot, Stream, Blur, and ParkPow. It does not contain the recognition models themselves.

Use the quickstart below to recognize a plate in an image, or jump to the repository map to find a tool for another workflow.

Image recognition quickstart

Prerequisites

• Python 3.10 or newer
• A Plate Recognizer API token, or a running self-hosted Snapshot SDK

Clone the repository and install the two dependencies required by the image client:

git clone https://github.com/parkpow/deep-license-plate-recognition.git
cd deep-license-plate-recognition
python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
python -m pip install requests pillow

Run recognition against the cloud API:

python plate_recognition.py --api-key MY_API_KEY /path/to/vehicle.jpg

Replace MY_API_KEY locally and do not commit your token. To use a running self-hosted Snapshot SDK instead:

python plate_recognition.py --sdk-url http://localhost:8080 /path/to/vehicle.jpg

The command returns JSON containing the recognized plate, confidence scores, and bounding box. The complete response schema is in the Snapshot API reference.

[
  {
    "version": 1,
    "results": [
      {
        "box": {"xmin": 85, "ymin": 85, "ymax": 211, "xmax": 331},
        "plate": "ABC123",
        "score": 0.904,
        "dscore": 0.92
      }
    ],
    "filename": "car.jpg"
  }
]

Common image workflows

Limit recognition to one or more regions:

python plate_recognition.py --api-key MY_API_KEY --regions fr --regions it /path/to/car.jpg

Process several files or a shell glob:

python plate_recognition.py --api-key MY_API_KEY /path/to/car1.jpg /path/to/car2.jpg /path/to/trucks*.jpg

Run python plate_recognition.py --help for all available output, annotation, cropping, and engine options. See the bulk-processing guide for a longer walkthrough.

Repository map

Each subproject has its own dependencies. Follow its README or requirements file instead of assuming the image-client installation covers the entire repository. The root pyproject.toml contains the shared Python development environment, managed with uv.

Development setup

Install uv, then create the root development environment from uv.lock:

uv sync --locked

The root is a collection of scripts rather than an installable package. Run tools in its environment with uv run, for example:

uv run pre-commit install
uv run pre-commit run --all-files

The first command installs the repository's Git hooks; the second runs every configured hook against the repository.

Plate redaction

number_plate_redaction.py detects plates—including small or barely readable ones—and can save a blurred copy of an image.

python number_plate_redaction.py --api-key MY_API_KEY vehicle.jpg --save-blurred

Useful options include:

• --split-image for high-resolution images; this uses three API lookups.
• --ignore-regexp REGEX to leave matching plates unblurred. Repeat the option to provide more than one expression.
• --ignore-no-bb to ignore results without a vehicle bounding box.

python number_plate_redaction.py \
  --sdk-url http://localhost:8080 \
  --split-image \
  --save-blurred \
  vehicle.jpg

Run python number_plate_redaction.py --help for the complete CLI reference.

FTP and SFTP processing

Use ftp_and_sftp_processor.py to fetch images from your own FTP or SFTP server and recognize them with the cloud API or a self-hosted SDK.

python -m pip install requests pillow paramiko
python ftp_and_sftp_processor.py \
  --api-key MY_API_KEY \
  --hostname FTP_HOST_NAME \
  --ftp-user FTP_USER \
  --ftp-password FTP_PASSWORD \
  --folder /path/to/server_folder

Add --protocol sftp for SFTP. Authentication can use --ftp-password or --pkey. The --delete option removes processed remote files, so test without it first. Run python ftp_and_sftp_processor.py --help for camera, polling, output format, and SDK options.

See the FTP/SFTP bulk-processing guide for setup details. Plate Recognizer also provides a hosted FTP integration.

Automatic image transfer

transfer.py watches a directory, recognizes new images, and moves processed images into an archive. It can optionally forward results to ParkPow.

python -m pip install requests watchdog jsonlines
python transfer.py --help

The help output includes examples for the cloud API and a self-hosted SDK, as well as the required source, archive, and camera-path options.

Language examples

• C++
• C#
• Java
• Android example in Java
• Android example in Kotlin
• Node-RED integration

For other languages, start with the API documentation or convert the API's curl examples with curlconverter.

Support and license

For product questions, contact Plate Recognizer. This repository is provided under the terms in LICENSE.

Plate Recognizer is a subsidiary of ParkPow.