# matthewearl/deep-anpr

Using neural networks to build an automatic number plate recognition system

Repository: https://github.com/matthewearl/deep-anpr
Canonical: https://ross.abutalabs.com/products/deep-anpr
Language: Python
License: MIT
License Family: permissive
Last push: 2019-11-14T22:40:30+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3775, "days_push": 2484, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1856, forks 685 (observed 2026-08-28T04:05:44.959968+00:00)

## What it is
An experimental Python project that uses neural networks (TensorFlow) to build an automatic number plate recognition (ANPR) system. It includes scripts for generating synthetic training data, training a CNN, and detecting license plates in images.

## Use cases
- recognize license plates in images with a neural network
- train a CNN for number plate detection
- generate synthetic UK number plate training images
- learn how ANPR systems work end to end
- detect number plates in a photo from the command line

## When to choose
- you want to learn or tinker with how CNN-based ANPR works
- you read the accompanying blog post and want a code companion
- you need a starting point to build a custom plate recognition system

## When to avoid
- you need a production-ready or accurate number plate recognition system
- you want a maintained project with active development
- you cannot spare significant GPU time and disk space for training data

## Facets
- artifact type: library
- maturity: experimental
- function: machine-learning, deep-learning, computer-vision, ocr, image-processing
- domain: computer-vision, machine-learning, deep-learning, image-processing
- platform: python
- tags: anpr, license-plate-recognition, neural-networks, tensorflow, opencv, synthetic-data, linux, macos, gpu

## Member repositories
- matthewearl/deep-anpr (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:44.959968+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:16:39.424804+00:00, confidence not recorded.
  - readme: https://github.com/matthewearl/deep-anpr (fetched 2026-08-28T04:05:44.959968+00:00, sha c6b7e8f8f11e)
- Data as of 2026-08-30T08:39:29.467469+00:00.