Door unlock for employees using image vision
Door unlock for employees using image vision
Topic: Door unlock for all participants in this hackathon using image/video/computer vision
Requirements
The doors should unlock for any authorized working at the entrance of the office.
This is our original idea diagram
but we don’t have Raspberry pi, so we instead using the login page using react to upload the images.
summary
1) There is an admin panel where only the admin can register an employee and upload images
2) Any person will come to the camera module and the camera module will take pictures of the person
3) If the person is matched with an employee then the door will open or it’ll not open if the person is not identified.
The Ui is made using React and the backend is on java spring boot
we using CNN to implement face recognition
we are using IOT client affnine_deltaleaf link) to connect door and camera
At first the admin will register the employee on the sign up form using employee id, First name, Last Name and image of the employee
then
It will hit the spring server to store all the details in the server’s local storage
then after that
it will pop an alert showing Registration is successful
Now
on Log in screen there is a button to take a picture of the employee and the picture will uploaded to the spring server
then
the spring server will call the AI function where using CNN the function will try to match the image with the existing employee images if there is a match the function will return a boolean value
which will hit the IOT server and then the door will open or close as per the request
| Employee | UnKnown Person | |
|---|---|---|
| Door | Open |
Closed |
```