Smart Authentication

Smart KYC server which takes Aadhaar Card images as the input and processes the data for the individual.

It uses some open source project to work properly :

• Tesseract - Tesseract Open Source OCR Engine
• Text-Detection-CTPN - Text detection mainly based on ctpn model in tensorflow

Note : This Project is tested and developed on Linux (Linux Mint) and Python 3.6

---

Installation

Install Tesseract Binary for ocr

• For Ubuntu and Ubuntu based Distros

sudo apt-get install tesseract-ocr

For other other OS users refer this link : Installing Tesseract for OCR

Install Python Dependencies

• Create a virtual environment with Conda

 conda env create --file environment.yml

Note : The name of environment in the environment.yml file is ml-old.

• Python Libraries required (Python 3.6)

easydict==1.9
tensorflow==1.10.0
numpy==1.16.4
Flask==1.1.2
Werkzeug==1.0.1
six==1.15.0
Keras==2.0.8
matplotlib==3.2.2
scipy==1.2.1
waitress==1.4.3
Cython==0.29.21
pytesseract==0.3.5
ipython==7.17.0
Pillow==7.2.0
PyYAML==5.3.1
scikit_learn==0.23.2

Files Description and Resources

Details About Files In This Directory

main.py :  File for starting the server
└───routes
│   └───/image/upload : processing the card image (ocr + face embeddings)
│   │
│   └───/live : Detect and Recognise face live (to be removed, video feed in separate window)
│   |   
│   └───/browse : video stream from webcam
│   |   
│   └───/details : renders the form page
│   |   
│   └───/feed : renders the page with video stream
│   |   
│   └───/ : home path, renders index page

utils.py : utils for MTCNN based face detection on Card

processing.py : utils for Aadhaar OCR

model.py : tensorflow configuration and model initiation

camera.py : code for in-browser video feed

requirements.txt : python Dependencies for this Project

environment.yml : yml file for the conda environment

Resources

Data and model files required to run this Project

data/ctpn.pb : ctpn.pb file

model/20170512-110547/ : Model folder

Download above two files and put them in the file struture as given below :

AuthX
└───data
│   └───ctpn.pb : ctpn.pb file
|
└───model
|   └───20170512-110547
|   |   └───20170512-110547.pb
|   |   └───model-20170512-110547.ckpt-250000.data-00000-of-00001
|   |   └───model-20170512-110547.ckpt-250000.index
|   |   └───model-20170512-110547.meta

Running the project

For Running the project, go to the project Directory and run the command below :

python main.py

this should start a server at port 5000

Tasks To Do

• Live video feed extraction
• Improve CSS/HTML
• Refactor Code
• In browser video feed