License plate recognition

Description

A Dutch plate recognition written in Python using only image processing techniques and no machine learning. The process involves:

• Plate localization
• Character segmentation
• Character recognition
• Adjusting the output

The system works well on videos of Dutch license plates up to about 10m distance.

Input

You run your video by putting under dataset/trainingvideo.avi.

Output

You can see the output under dataset/Output.csv.

Project setup

The shell script evaluator.sh is used for running the project and calculating scores.
This file initially runs your main.py file, followed by evaluation.py. Do not modify either one of evaluator.sh or evaluation.py if you want to see proper outputs.

Rest of the project:

• CaptureFrame_Process.py for reading the input video
• Localization.py for figuring out the location of the plate in a frame
• Recognize.py for figuring out what characters are in a plate
• helpers/ additional methods to help you get started (you do not have to use them, and are there for inspiration)
• requirements.txt if you want to use additional Python packages make sure to add them here as well