License Plate Recognition with Keras
This is a port of Matthew Earl's deep ANPR CNN to Keras, allowing easier experimentation with new architectures and training methods.
Getting Started
./gen.py 1000: Generate 1000 test set images in test/(test/ must not already exist)
./train.py: Train the model. A GPU is recommended for this step. You can uncomment line 63 if you want to display the predictions for each test image while training.
Prerequisites
For the image generation process, it is required:
- Background images at
bgs/folder. Background images to be used for generating training/test images are included in this repository. However, you can extract ~3GB of background images from the SUN database into bgs/ with./extractbgs.py SUN397.tar.gz(bgs/ must not already exist.) The tar file (36GB) can be downloaded here. This step is optional and may take a while as it will extract 108,634 images. - At least one .ttf font to be in the fonts/ directory. The Mandatory Font is already included, but you may add other fonts.
The project has the following dependencies:
Numpy
Tensorflow
Keras
OpenCV