tracker

Overview

The aim of the project is to provide a face recognition based attendence system. The first part of the project is a Python module used as a library to provide face recognition functionality; managing system users and their photos, training the model and making predictions. The second part is a Web interface to provide the configuration and administration of the system.

The recognition module

The __init__.py of the module contains the face_cascade object. This is a common ressource to be used in order to detect the face from the picture.

The trainer.py contains the Trainer class which is the responsible of reading the photos to train them in a model. The system uses the three available algorithms in OpenCV (LBPH, FisherFace and EigenFace) and store three trained models in three files.

The recognizer.py contains the Recognizer which use the trained models to make predictions.

The capture.py contains a function to capture photos of a certain user and store them in the proper way.

The django web interface

• The web interface consists of the following functionalities:
• Login and Homepage showing last stats of the system
• Add user page to add new system users
• Capture page to capture users' photos from a local or remote device
• Train page to train the models
• Attendence page to view the attendence records
• A RESTful API interface to send attendence records (from the raspberry pi in the prototype device)

Screenshot

Note

This project is a part of a complete enterprise automation system that includes IoT, NLP, and AI functionalities ending up with an intelligent automation system.

Requirements

You need Python3.6 to be installed on your system with some additional libraries indicated in requirements.txt.

You can install them with pip3 tool using the following command: pip install -r requirements.txt

You also need to create an empty postgresql database for django (or use a different DBMS).

Running

To run the project you need first to initialize the database from django:

• python3 manage.py makemigrations
• python3 manage.py migrate
• python3 manage.py createsuperuser

Then start the server with: python3 manage.py runserver