Face recognition concierge to announce new visitors

This tutorial shows you how to train a facial recognition model to identify visitors and send an email announcing their arrival.

Tutorial Requirements

• Python version 3
• A webcam (your laptop’s built-in webcam or an external one)
• A free Twilio SendGrid account to send up to 100 free emails per day

Step 0: Clone repo and install dependencies

Clone this example project, and change into the directory from the command line.

$ git clone git@github.com:loopDelicious/facial-recognition.git
$ cd facial-recognition

Create a virtual environment called venv. Activate the virtual environment, and then install the required Python packages inside the virtual environment. If you’re on Unix or Mac operating systems, enter these commands in a terminal.

$ python -m venv venv
$ source venv/bin/activate
(venv) $ pip install -r requirements.txt

If you’re on a Windows machine, enter these commands in a command prompt window.

$ python -m venv venv
$ venv\Scripts\activate
(venv) $ pip install -r requirements.txt

Step 1: Create a custom face recognition dataset

Create a new subfolder inside the dataset directory using your first name, like Joyce, to contain your photos.

(venv) $ python headshots.py Joyce

Then run this command to open a new webcam window, passing in the name of your new subfolder. Use headshots_picam.py if using a Pi camera. Press the spacebar to take at least 10 pictures of your face from different angles. When you're done, ESC to close the window. Repeat this step to add more friends, creating a separate folder for each person.

Step 2: Train the model

(venv) $ python encode_faces.py

Run this command to analyze the photos and output a new file encodings.pickle that contains criteria for identifying these faces.

Step 3: Test the model

(venv) $ python facial_req.py

Run this command to open a new webcam window. If your face is highlighted with a yellow box alongside your name, the model has been properly trained. Hit q to quit the program.

Step 4: Set up SendGrid email notifications

Create a new file called .env (notice the dot in front of the filename), formatted like .env.example. Save your API key from the SendGrid settings and other configuration details in this file.

(venv) $ python send_test_email.py

Run this command to send a test email.

Step 5: Add email notifications to facial recognition

(venv) $ python facial_req_email.py

Run this command to open a new webcam window and try it out. If someone from your dataset is recognized, the webcam will snap a photo and send an email notification to announce the new arrival.

---

Attributions

Forked from this Raspberry Pi 4 Facial Recognition tutorial

https://www.tomshardware.com/how-to/raspberry-pi-facial-recognition

Included code samples from these Face Recognition tutorials

https://www.pyimagesearch.com/2018/06/11/how-to-build-a-custom-face-recognition-dataset/ https://www.pyimagesearch.com/2018/06/18/face-recognition-with-opencv-python-and-deep-learning/