id-matching

A face recognition application that takes a photo and checks for a match in a folder of identity documents.

Usage

py idMatch.py <input_image> <documents_folder>

Examples

py .\idMatch.py .\Test_Data\sample2.png .\Test_Data\documentfolder\
.\Test_Data\sample2.png  not matched in  .\Test_Data\documentfolder\

py .\idMatch.py .\Test_Data\sample1.png .\Test_Data\documentfolder\
.\Test_Data\sample1.png  matched  .\Test_Data\documentfolder\document1.png  in  .\Test_Data\documentfolder\

Requirements

Use the package manager pip to install the following packages.

pip install tensorflow

pip install Keras

pip install opencv-python

pip install mtcnn

pip install matplotlib

pip install keras_vggface

Description

The executable named idMatch accepts as parameters:

1. The path to the input image.
2. A path to a folder containing identity documents.

The user can type in a command line “py idMatch <input_image> <documents_folder>” and receive as output: e.g. “photo1 matched document1 in folder1” if the code matched photo1 to document1 in folder1 or “photo1 not matched in folder1”.

References

1. MTCNN face detection implementation for TensorFlow
2. VGGFace implementation with Keras