I am using RaspberryPi to hold the ML model I build and it will work as a Security Door Lock
-
Face Recognition
- Using Tensorflow Building a customized Siamese Model for One-Shot Learning
- Using Roboflow and Google Colab to gather Anchor, Positive for User(Me & My family). Negative using lfw dataset
- Recognize my or my family's faces or identify as a Stranger
-
Object Detection
- Using Yolov5 with customized dataset created in Roboflow
- Detect nearby objects after Face Recognition failed to identify a user in database
- Identify objects like packages, presents, or dangerous objects to send a relative message to my Phone
-
RaspberryPi 4B
- Using RaspberryPi to hold both ML models
- Attach with a Camera Module
- Attach with a Servo to lock the door
-
3D printing
- Using Solidworks to CAD the case for holding a 15000wAh battery charger, RaspberryPi and the Camera
- 3D printing the STL files in School
- 3D print a gimbal and attach a better camera
- Use bluetooth module in RaspberryPi to recieve data transmitted from camera
- Use Actuators to lock the door better