SMART DOOR

FaceLock is an automated door locking and unlocking system powered by facial recognition. Designed for Raspberry Pi 5, it uses computer vision to identify authorized users and control a door lock accordingly. Ideal for home automation, secure offices, or smart lockers.

FEATURES

• Real-time facial recognition
• Automatic door locking/unlocking
• Camera and DC motor integration with Raspberry Pi
• Local facial data storage and processing
• Easy setup and deployment

HARDWARE REQUIREMENTS

• Raspberry Pi 5
• Pi Camera
• DC motor
• L298N motor driver
• Display

Proper WIRING for the project is available below.

USAGE

1. Project Setup

Use the following command to download the project and its requirements:

git clone https://github.com/Nikhil-Chanda/smart-door.git
cd smart-door
pip install -r requirements.txt

Make sure the Pi camera is connected properly and enabled in the settings (sudo raspi-config > Interface Options > Camera > Enable).

2. Collect Data

To capture images for training the facial recognition model, Run:

python3 image_capture.py

• A live camera feed will open.
• When your face is detected (indicated by a green box), press "SPACE" key to capture an image.
• Press "q" to exit after you're done.

Note: It is recommended to capture in total of atleast 9 different directions per person

↖ ↑ ↗
← O →
↙ ↓ ↘

3. Train Model

To train the model, run

python3 model_training.py

• This script processes the captured images and generates facial encodings.
• These encodings are saved and used during live recognition.
• The training process is lightweight and optimized to run smoothly on Raspberry Pi 5.

4. Running the System

To start the real-time facial recognition and motor control system, Run:

python3 main.py

• The camera continuously scans for faces.
• If a match is found in the trained dataset, the motor driver is activated to unlock the door.
• The system re-locks the door after a brief delay or based on further logic.

WIRING

Note We refer to the BOARD layout in this project and not BCM layout. That means pin 1 will be 3v3 power, pin 2 will be 5v power and so on.

Raspberry Pi 5 GPIO

L298N Motor driver

• Connections are as follows:
  • RPi pin 2 -> L298n 5v
  • RPi pin 6 -> L298n GND
  • External 12v Power source -> L298n VCC
  • RPi pin 12 -> L298n IN1
  • RPi pin 16 -> L298n IN2
  • RPi pin 38 -> L298n ENA
  • DC Motor -> OUT1 & OUT2
• To close the motor:
  • RPi pin 4 -> RPi pin 37