Face Recognition with Python [source code included]

Python can detect and recognize your face from an image or video

Face Detection and Recognition is one of the areas of computer vision where the research actively happens.

The applications of Face Recognition include Face Unlock, Security and Defense, etc. Doctors and healthcare officials use face recognition to access the medical records and history of patients and better diagnose diseases.

About Python Face Recognition

In this python project, we are going to build a machine learning model that recognizes the persons from an image. We use the face_recognition API and OpenCV in our project.

Tools and Libraries

• Python – 3.x
• cv2 – 4.5.2
• numpy – 1.20.3
• face_recognition – 1.3.0

To install the above packages, use the following command.

pip install numpy opencv-python

To install the face_recognition, install the dlib package first.

pip install dlib

Now, install face_recognition module using the below command

pip install face_recognition

Download Face Recognition Python Code

Please download the source code of python face recognition project: Face Recognition Project Code

Project Dataset

We can do this face recognition project using our own dataset. For this project, let’s take the cast of the popular American web series “Friends” as the dataset. The dataset is included with face recognition project code, which you downloaded in the previous section.

Steps to develop face recognition model

Before moving on, let’s know what face recognition and detection are.

Face recognition is the process of identifying or verifying a person’s face from photos and video frames.

It is defined as the process of locating and extracting faces (location and size) in an image for use by a face detection algorithm.

Face recognition method is used to locate features in the image that are uniquely specified. The facial picture has already been removed, cropped, scaled, and converted to grayscale in most cases. Face recognition involves 3 steps: face detection, feature extraction, face recognition.

OpenCV is an open-source library written in C++. It contains the implementation of various algorithms and deep neural networks used for computer vision tasks.

1. Prepare the dataset

Create 2 directories, train and test. Pick an image for each of the cast from the internet and download it onto our “train” directory. Make sure that the images you’ve selected show the features of the face well enough for the classifier.

For testing the model, let’s take a picture containing all of the cast and place it onto our “test” directory.

For your comfort, we have added training and testing data with the project code.

2. Train the model

First import the necessary modules.

import face_recognition as fr
import cv2

import numpy as np
import os

The face_recognition library contains the implementation of the various utilities that help in the process of face recognition.

Now, create 2 lists that store the names of the images (persons) and their respective face encodings.

path = "./train/"

known_names = []
known_name_encodings = []

images = os.listdir(path)

Face encoding is a vector of values representing the important measurements between distinguishing features of a face like the distance between the eyes, the width of the forehead, etc.

We loop through each of the images in our train directory, extract the name of the person in the image, calculate its face encoding vector and store the information in the respective lists.

for _ in images:
image = fr.load_image_file(path + _)
image_path = path + _
encoding = fr.face_encodings(image)[0]

known_name_encodings.append(encoding)
known_names.append(os.path.splitext(os.path.basename(image_path))[0].capitalize())

3. Test the model on the test dataset

As mentioned above, our test dataset only contains 1 image with all of the persons in it.

Read the test image using the cv2 imread() method.

test_image = "./test/test.jpg"

image = cv2.imread(test_image)

The face_recognition library provides a useful method called face_locations() which locates the coordinates (left, bottom, right, top) of every face detected in the image. Using those location values we can easily find the face encodings.

face_locations = fr.face_locations(image)

face_encodings = fr.face_encodings(image, face_locations)

We loop through each of the face locations and its encoding found in the image. Then we compare this encoding with the encodings of the faces from the “train” dataset.

Then calculate the facial distance meaning that we calculate the similarity between the encoding of the test image and that of the train images. Now, we pick the minimum valued distance from it indicating that this face of the test image is one of the persons from the training dataset.

Now, draw a rectangle with the face location coordinates using the methods from the cv2 module.

for (top, right, bottom, left), face_encoding in zip(face_locations, face_encodings):
   matches = fr.compare_faces(known_name_encodings, face_encoding)
   name = ""

   face_distances = fr.face_distance(known_name_encodings, face_encoding)
   best_match = np.argmin(face_distances)

   if matches[best_match]:
       name = known_names[best_match]

   cv2.rectangle(image, (left, top), (right, bottom), (0, 0, 255), 2)
   cv2.rectangle(image, (left, bottom - 15), (right, bottom), (0, 0, 255), cv2.FILLED)

   font = cv2.FONT_HERSHEY_DUPLEX
   cv2.putText(image, name, (left + 6, bottom - 6), font, 1.0, (255, 255, 255), 1)

Display the image using the imshow() method of the cv2 module.

cv2.imshow("Result", image)

Save the image to our current working directory using the imwrite() method.

cv2.imwrite("./output.jpg", image)

Release the resources that weren’t deallocated(if any).

cv2.waitKey(0)
cv2.destroyAllWindows()

Python Face Recognition Output

Let’s see the output of the model.

Summary

Face recognition means finding and matching faces from images or videos. It’s used in mobile phones, security systems, and smart apps. In this project, we use Python and OpenCV to build a real-time face recognition system. It can detect faces using a webcam and match them with known people using a face database.

In this machine learning project, we developed a face recognition model in python and opencv using our own custom dataset.

This project helps beginners learn how face detection and recognition work. It teaches concepts like facial landmarks, encodings, image comparison, and OpenCV basics. It’s perfect for real-time systems and makes a strong resume project. You can even build an attendance app or secure login system using this idea.