Face-Recognition-Door-Lock-with-AWS-Rekognition-Raspberry-Pi3

Face Recognition Door Lock with AWS Rekognition & Raspberry Pi3 it works with RPI3 using the camera module

Video

Hardware Requirements:

1.Raspberry Pi (Any Version Will Work)

2.Raspberry Pi Camera (Also USB Webcam Can Be Used)

3.Push Button

4.Electric Door Lock

Software Dependencies:

Python2

Boto3

pip install boto3

Python3

Boto3

pip3 install boto3

Step 1 : Create a AWS S3 Bucket in that bucket create folders with the name of the students and add their images atleat 5-10

Step 2 : Go to IAM and create a new user and set access type to Programmatic access

Step 3 : Set permissions for S3 and Rekoognition to full access

Step 4 : Complete the process you will get Accesss Key ID & Secret Access Key Copy both and add it in train.py and main.py

train.py

s3_client = boto3.client(
    's3',
    aws_access_key_id='',# add the aws access key
    aws_secret_access_key=''# add the aws secret access key
    
)

collectionId='' #collection name

rek_client=boto3.client('rekognition',
                            aws_access_key_id='',# add the aws access key
                            aws_secret_access_key='',# add the aws secret access key
                            region_name='',)# add the region here

recognition.py

rek_client=boto3.client('rekognition',
                        aws_access_key_id='',# add the aws access key
                        aws_secret_access_key='',# add the aws secret access key
                        region_name='ap-south-1',)# add the region here

Step 5 : Add the S3 Bucket Name & Folder to save the images on pi

Both files

bucket = '' #S3 bucket name

recognition.py

directory = '' #folder name on your raspberry pi

Run

First Run Train.py File on RPI

python train.py

Run main.py File on RPI ,connect the booton with GPIO 26

python main.py

License & Copyright

© Arbaz Khan

Licensed under the MIT License