OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Image & Video AI

End To End For Chinese Plate Recognition

A Python computer-vision application that locates and rectifies Chinese license plates in images, then uses a convolutional neural network to recognize them. Its pipeline combines U-Net segmentation and OpenCV edge detection with TensorFlow and Keras models.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

End To End For Chinese Plate Recognition — OpenViST project profile in Image & Video AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when evaluating a Python-based pipeline that handles plate localization and image correction before neural-network recognition, or when considering recognition for image inputs from a vehicle-camera workflow.

Who it is for

  • Developers exploring image-based Chinese license-plate recognition.
  • Teams considering a plate-image recognition component for a vehicle-camera or access-control workflow.

Documented uses

  • Locate a plate region within an image.
  • Correct the plate image before recognition.
  • Recognize Chinese license-plate characters from an image.
Before you choose

Requirements & limits

The repository is public and lists an Apache-2.0 license.

  • The listed environment is Python 3.6, TensorFlow 1.15.2, OpenCV 4.1.0.25 and Keras 2.3.1.
  • For a complete plate image that does not need localization, the README advises dimensions of 240 × 80 pixels or smaller; larger images may trigger localization and reduce recognition quality.
  • The README reports results in tilted-angle, bright and dim conditions; these are project claims rather than independent performance validation.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Prepare the documented runtime for plate recognition

Set up the stated Python and machine-learning library environment before using the recognition pipeline.

Before you start

  • Python 3.6
  • TensorFlow 1.15.2
  • OpenCV 4.1.0.25
  • Keras 2.3.1

Steps

  1. Use Python 3.6.

  2. Prepare TensorFlow 1.15.2, OpenCV 4.1.0.25, and Keras 2.3.1.

  3. Consult the project README for the documented pipeline stages before proceeding: U-Net segmentation, OpenCV edge detection and plate correction, then CNN recognition.

Expected outcome

The documented runtime versions are identified, and the README describes the pipeline stages that follow setup.

Guide limitations

  • The README does not provide installation commands or steps for running the application.
  • For a complete plate image that does not need localization, use an image no larger than 240 × 80; larger images may be treated as containing additional regions and trigger localization.
Documentation 1 ↗Documentation 2 ↗
Evidence and transparency

Sources & review

Official repository ↗Guide documentation ↗
Source record details
Source availability
Active
Sources checked
15 October 2026
Original website title
duanshengliu/End-to-end-for-chinese-plate-recognition · Official homepage title · retained in its source language
How we review sourcesSuggest a correction