How do you build a face recognition door entry on Raspberry Pi?
A Pi camera recognises enrolled faces with OpenCV and the face_recognition library, and opens a door lock only for them. You need Raspberry Pi 5, Raspberry Pi 27 W USB-C power supply, SanDisk microSD card 32 GB (A1, Class 10), Raspberry Pi Camera Module 3, Raspberry Pi 5 to Camera cable (22 to 15 pin), Relay Module 5V, Solenoid lock 12 V, Rectifier Diode 1N4007 and 12 V 2 A DC adapter.
- Level
- Advanced
- Time
- About 360 min
- Parts
- 9
Run face detection and recognition on a Pi and act on the match distance.
Parts list
- Single Board Computers
- SBC Accessories
- SBC Accessories
- SBC Accessories
- Cables & Wires
- Interfaces
- Motors & Actuators
- Semiconductors
- Power
How to build it
- Install OpenCV and face_recognition in a virtual environment
- Take 10 photos of each person and save their encodings
- Detect faces in the live camera frame and compare encodings
- Open the lock when the distance is below 0.5
- Keep a manual key: face systems can be fooled by photos
Common questions
What parts do I need for the face recognition door entry on Raspberry Pi?
Raspberry Pi 5, Raspberry Pi 27 W USB-C power supply, SanDisk microSD card 32 GB (A1, Class 10), Raspberry Pi Camera Module 3, Raspberry Pi 5 to Camera cable (22 to 15 pin), Relay Module 5V, Solenoid lock 12 V, Rectifier Diode 1N4007 and 12 V 2 A DC adapter.
How long does the face recognition door entry on Raspberry Pi take to build?
About 360 minutes for a advanced maker.
