/Face-Recognizer

Thesis: Face Recognizer

Primary LanguagePythonGNU General Public License v3.0GPL-3.0

Facial Recognition System Designed for School, Business Environment

This is my Thesis at the University for the final BSc semester.


Thesis

Read my Thesis here

Setup

Required

  • OpenCV - 2.4.
  • Python - 2.7.
  • xlsxwriter package (pip install xlsxwriter)
  • pandas package (pip install pandas)

Install OpenCV as you can see on the offical site. Or you can use Anaconda enviroment for easy setup.

After the setup you should check settings_for_recognition.json because there you can see the global settings.

Run it

  1. Collect data and place it in input_images folder if you would like to prepare the data from that source. If you would like to use webcam than just skip this step.
  2. Run python face_recognizer_menu.py
  3. Choose from the menu points:
    • 1: Prepare the training data from the folder (input_images)
    • 2: Prepare training data from webcam (results will be saved to output_images)
    • 3: Train the face recognizer with the prepared data (model will be saved to saved_model)
    • 4: Test face recognition with a webcam
    • 5: Recognize from camera and create attendance sheet
    • 6: About
    • 7: Exit from the application

Folder Structure

cascades/
	hc_face.xml
input_images/
	It can be empty if you prepare data with a webcamera
	Peter/
		peter1.jpg
		peter2.jpg
		...
	Dori/
		dori1.jpg
		dori2.jpg
		...
	Mona/
		mona1.jpg
		mona2.jpg
		...
	...
output_images/
	There are generated folders and images for the training
saved_models/
	Here you can saved the trained model
documentation/
	Face_Detection_And_Recognition_By_Gabor_Vecsei.pdf

About

Gábor Vecsei

2016.12.09.