CCPD: Chinese City Parking Dataset for License Plate Recognition

CCPD is a large-scale image dataset for license plate detection and recognition, released by detectRecog for the ECCV 2018 conference. It provides annotations for plate bounding boxes and character sequences across diverse real-world parking environments.

Use Cases

• Training object detection models to predict bounding boxes for license plates
• Developing OCR systems to recognize Chinese characters and alphanumeric sequences
• Benchmarking model performance against environmental factors like tilt and illumination

Strengths

• ECCV 2018 peer-reviewed publication
• MIT licensed for open use
• Includes diverse environmental conditions like tilt and weather

Limitations

• Geographic bias toward Chinese license plate formats
• Context restricted to parking scenarios

Provenance

Source

ECCV 2018 publication by detectRecog

Collection Method

annotated

Freshness

Last updated February 2024; originally published in 2018.

Geography

China