AdivarekarBhumit / ID-Card-Segmentation

Segmentation of ID Cards using Semantic Segmentation

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

ID-Card-Segmentation

Segmentation of ID Cards using U-Net

U-Net Architecture

U-net

Our Result's

Test_Image Output_Image

Requirements

  • Tensorflow-GPU 1.12
  • Keras 2.1
  • OpenCV 3.4.5
  • Numpy 1.16

Dataset

  • Download Dataset
python dataset/download_dataset.py
  • Combine To single npy file (First Download the dataset)
python dataset/stack_npy.py

Train Model

  • Start Training
python model/train.py

Training data in 100 epochs. This data was trained on google colab

Test Model

python test_model.py

Benchmarks

IoU Loss

IoU Loss

Binary Accuracy

Binary Accuracy

Val IoU Loss

Val.IoU Loss

Val Binary Loss

Val.Binary Accuracy

Citation

Please cite this paper, if using midv dataset, link for dataset provided in paper

@article{DBLP:journals/corr/abs-1807-05786,
  author    = {Vladimir V. Arlazarov and
               Konstantin Bulatov and
               Timofey S. Chernov and
               Vladimir L. Arlazarov},
  title     = {{MIDV-500:} {A} Dataset for Identity Documents Analysis and Recognition
               on Mobile Devices in Video Stream},
  journal   = {CoRR},
  volume    = {abs/1807.05786},
  year      = {2018},
  url       = {http://arxiv.org/abs/1807.05786},
  archivePrefix = {arXiv},
  eprint    = {1807.05786},
  timestamp = {Mon, 13 Aug 2018 16:46:35 +0200},
  biburl    = {https://dblp.org/rec/bib/journals/corr/abs-1807-05786},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}

About

Segmentation of ID Cards using Semantic Segmentation

License:MIT License


Languages

Language:Python 100.0%