Face Recognition Door Lock System using Raspberry Pi

The advancements in Face Recognition technology have led to its widespread use in surveillance and security applications. In this tutorial, we will explore how to create a Face Recognition Door Lock System using Raspberry Pi. The project consists of three main phases:

1. Data Gathering: We will collect authorized face samples that will be used to open the lock.
2. Training the Recognizer: The collected face samples will be used to train the Face Recognizer.
3. Face Recognition: The trained Recognizer will be utilized to recognize faces. If Raspberry Pi successfully recognizes a face, it will unlock the door.

Paraphrased: “In this project, we will utilize a Raspberry Pi 3, a solenoid lock, and a Pi camera to construct a face recognition-based door lock system. Previously, we have explored the implementation of a solenoid lock with Raspberry Pi, as well as developed various projects incorporating the Pi camera, such as the Web Controlled Raspberry Pi Surveillance Robot, IoT-based Smart Wi-Fi doorbell, and Smart CCTV Surveillance System.”

Components Required

Paraphrased: “To build the face recognition-based door lock system, you will need the following components:
– Raspberry Pi 3 (any version)
– Solenoid Lock
– Relay Module
– Jumper Wires”

Circuit Diagram

Circuit diagram for Face Recognition Door Lock using Raspberry Pi is given below.

To connect the Raspberry Pi and Solenoid Lock, we use a relay module. The Solenoid Lock requires a higher voltage (9-12V) than the Raspberry Pi can provide (5V), so a 12V adapter is used to power the Solenoid Lock. The VCC and GND pins of the relay module are connected to the 5V and GND pins of the Raspberry Pi. The input pin of the relay is connected to GPIO23 of the Raspberry Pi.

The positive pin of the Solenoid Lock is connected to the positive rail of the 12V adapter, while the negative pin of the Solenoid Lock is connected to the Common (COM) terminal of the relay. The Normally Open (NO) terminal of the relay is connected to the negative terminal of the 12V adapter.

The breadboard setup for Raspberry Pi 3 Face Recognition Door Lock is shown below:

Installing OpenCV in Raspberry Pi 3

Here OpenCV library is used to detect and recognize faces. To install the OpenCV, first, update the Raspberry Pi.

sudo apt-get update

Then use the following commands to install the required dependencies for installing OpenCV on your Raspberry Pi.

sudo apt-get install libhdf5-dev -y 
sudo apt-get install libhdf5-serial-dev –y 
sudo apt-get install libatlas-base-dev –y 
sudo apt-get install libjasper-dev -y 
sudo apt-get install libqtgui4 –y
sudo apt-get install libqt4-test –y

After that, use the below command to install the OpenCV on your Raspberry Pi.

pip3 install opencv-contrib-python==4.1.0.25

Installing other Required Packages for Face Recognition

Before we start programming the Face Recognition Door Lock System using Raspberry Pi, we need to install the necessary packages.

To install dlib, a modern toolkit that includes Machine Learning algorithms and tools for real-world problems, use the following command.

pip3 install dlib

Installing face_recognition module:The face recognition library is used to recognize and manipulate faces from Python through the command line. To install this library, use the following command.

Pip3 install face_recognition

The imutils library is used to simplify essential image processing functions, such as translation, rotation, resizing, skeletonization, and displaying Matplotlib images, when working with OpenCV. To install the imutils library, execute the following command.

pip3 install imutils

Installing pillow: Pillow is used to open, manipulate, and save images in a different format. Use the below command to install pillow:

pip3 install pillow

Programming for Raspberry Pi Face Recognition Door Lock

As mentioned previously, this project is divided into three phases: data gathering, training the Recognizer, and face recognition. The programs for each phase are provided at the end, and we will now explain each of them in detail.

1. Data Gathering

During the initial phase of the project, you need to create a dataset to store the faces with different IDs. To accomplish this, start by creating a project directory where all the project data will be saved.

mkdir FaceRecognition

Besides the three python program and Dataset, this directory also has a Facial Classifier file. All three python programs and Facial Classifier file is given at the end of the page.

Within the FaceRecognitionProject directory, create a new subdirectory called Dataset to store the face samples.

mkdir Dataset

Next, open a Nano editor file in the FaceRecognitionProject directory and copy-paste the data gathering program provided at the end.

sudo nano dataset.py

Data Gathering program is explained below:

Initialize the face detector by using the facial classifier file.

face_detector = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')

Now provide a user input command so that the user can enter the numeric face id before gathering the data.

face_id = input('\n enter user id end press ENTER ==>  ')

Inside the while loop, use the detector to extract the faces.

ret, img = cam.read()
img = cv2.flip(img, -1) # flip video image vertically
  gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
  faces = face_detector.detectMultiScale(gray, 1.3, 5)

Subsequently, save each captured frame as a file in the “dataset” directory, associating it with the corresponding person ID.”

cv2.imwrite("dataset/User." + str(face_id) + '.' + str(count) + ".jpg", gray[y:y+h,x:x+w])

Execute the Python scripts and input the face ID. Once a face is detected, the program will begin capturing samples and save them in the Dataset directory.

This is how the directories and program files will be structured:

2. Training the Recognizer

Paraphrased: “After collecting the face samples, proceed to train the Recognizer using these samples to enhance its accuracy in predicting faces.

Open a Nano editor file in the FaceRecognition directory, paste the trainer code provided (found at the end), and save it as trainer.py.

The Python script for training the Recognizer is described below:

Begin the code by importing all the necessary library files.”

import cv2
import numpy as np
from PIL import Image
import os

After that, enter the path where you saved the face samples.

path = 'dataset'

Subsequently, utilize the haarcascade_frontalface_default.xml facial classifier file to detect faces in the sample images. Then, employ the recognizer variable to instantiate an LBPH (Local Binary Pattern Histogram) Face Recognizer.

detector = cv2.CascadeClassifier("haarcascade_frontalface_default.xml");
recognizer = cv2.face.LBPHFaceRecognizer_create()

Now navigate to the directory containing the face samples using the previously initialized path.

imagePaths = [os.path.join(path,f) for f in os.listdir(path)]

After that, create two lists for storing face samples and IDs.

faceSamples=[]
ids = []

Convert the image samples to grayscale. Then, convert the PIL image to a numpy image.

PIL_img = Image.open(imagePath).convert('L') # convert it to grayscale
img_numpy = np.array(PIL_img,'uint8')

To extract the ID from the image path in the Dataset directory, we will split the path and retrieve the User ID and sample number.

id = int(os.path.split(imagePath)[-1].split(".")[1])

Now call the Faces and IDs list and feed them into trainer file.

faces,ids = getImagesAndLabels(path)
recognizer.train(faces, np.array(ids)

3. Recognizer

In the final step of the project, we will utilize face recognition technology to recognize faces from a live video feed. When Raspberry Pi identifies a recognized face, it will activate the relay module to open the solenoid lock.

The complete program for face recognition is provided at the end of the page. Here are some important aspects of the code:

The program uses the same library files and classifier file as the trainer program.

An array is used to associate names with each face ID.

names = ['None', 'Ashish', 'Thor']

Now capture video feed from the Raspberry Pi camera with a resolution of 640×480. If multiple cameras are being used, replace zero with one in the line cam = cv2.VideoCapture(0).

cam = cv2.VideoCapture(0)
cam.set(3, 640) # set video widht
cam.set(4, 480) # set video height

After that, inside the while loop, break the video into images and then convert it to grayscale.

ret, img =cam.read()
    img = cv2.flip(img, -1) # Flip vertically
   gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)

Use recognizer.predict function to check how much the face matches with the samples.

cv2.rectangle(img, (x,y), (x+w,y+h), (0,255,0), 2)
id, confidence = recognizer.predict(gray[y:y+h,x:x+w])

If confidence is less than 100, then open the lock. 0 means the perfect match.

if (confidence < 100):
            id = names[id]
            confidence = "  {0}%".format(round(100 - confidence))
            GPIO.output (relay, 1)

Testing the Raspberry Pi Face Recognition Door Lock

To execute these Python codes, you will need either an external monitor or a virtual monitor like VNC viewer. Run the dataset program to collect the face samples. Upon running this program, a window will appear where you can enter the ID number and press Enter. Subsequently, another window will appear to capture the face samples using the Raspberry Pi camera.

Next, run the trainer program. Once it successfully executes, it will create a ‘trainer.yml' file in your project directory. This file will be utilized by the Recognizer to recognize faces.

Finally, run the recognizer program. If a face is recognized in the video feed, you will see a box around the face with the person's name displayed, as shown below: