Skip to content

Latest commit

 

History

History
76 lines (53 loc) · 1.61 KB

File metadata and controls

76 lines (53 loc) · 1.61 KB

Face-Recognition-Based-Entry

This model recognizes face and marks and uploads entry on a firebase database (cloud firestore).

A program to take entry of a person using face recognition.

Getting Started

Use GUI.py to run.

Prerequisites

What things you need to run the software.

To install modules run:

pip install -r requirements.txt

Modules which will be installed includes:

face_recognition
opencv-python
pandas

Instructions

The major program files being used are:

    1.Registation.py
    2.Taking Entry.py

1.Registeration.py

This program takes id and name as input,then takes 2 images of the user and save the encodings along with details. Note:The code for taking images is selfie.py which is imported in this program.

For example:

If inputs are :

Id=1
Name=vineet

Then it will take 2 images of vineet, and save ([encodings of both images],[id two times],[name two times]) as a tuple and store it in a data.dat file.

example tuple:

(array[[encoding of 1st image][encoding of 2nd image]],array[1,1],array["vineet","vineet"])

Two images are used to increase accuracy.

You can check whats inside the data.dat file using testing_reading.py file (optional).

2.Taking Entry.py

This program uses data.dat file to load encodings and details. Use those encodings to match faces from webcam. If matched, it is supposed to take the attendance of that person if he's/she's in the database (i.e. data.dat file), or it will show unknown and close the window after a few seconds.

Acknowledgments

Team members:

Peeyush(Leader)
Nitish
Razaul
Vineet