hyperlpr-train_e2e

A simple code for creating licence plate images and train e2e network based on HyperLPR

Author	LCorleone
E-mail	lcorleone@foxmail.com

Requirements

run create_train_data.py to create plate image and corresponding labels. This repository also contains the plate generator and can generate thousands of plates.
reset the train data path and run train_nn.py to train your model.

The image size created automatically is 120 * 30, fix the input size when you use the e2e network. You can create and train your own e2e network if you want.
Generate at least 50000 images for training, less may degrade the performance.
Also, when tested in real scene, the e2e network performs not very well due to that the images' quality created automatically are still poor. If you have real image dataset and labels, it may be perfect.

Name		Name	Last commit message	Last commit date
Latest commit History 27 Commits
NoPlates		NoPlates
__pycache__		__pycache__
data/train_data		data/train_data
font		font
images		images
plate		plate
plate_train		plate_train
recognize_samples		recognize_samples
.gitignore		.gitignore
PlateCommon.py		PlateCommon.py
README.md		README.md
createImage_data.py		createImage_data.py
create_train_data.py		create_train_data.py
genplate.py		genplate.py
genplate_advanced.py		genplate_advanced.py
genplate_scene.py		genplate_scene.py
plate_chars_set.txt		plate_chars_set.txt
plate_size_china.md		plate_size_china.md
test_nn.py		test_nn.py
train_nn.py		train_nn.py