Skip to content

Latest commit

 

History

History
137 lines (100 loc) · 4.66 KB

README.md

File metadata and controls

137 lines (100 loc) · 4.66 KB
Raw

🦉 Baby Monitor

I'm having a baby, so I built a video baby monitor. It's a Raspberry Pi streaming video to a hosted web application, so I can conveniently access it from anywhere.

Login screenshot Day screenshot Night screenshot

I didn't like the idea of using any off-the-shelf baby monitor/surveillance camera connected to the internet. Closed circuit cameras don't offer the same convenience, and there is still the risk of some form of logging and/or telemetry if the camera needs to be on WiFi.

I put some effort into making my baby monitor private and secure. The video feed uses TLS, and any data being sent to the hosted web application requires an access token. Accessing the frontend of the application requires a simple username and password.

Hardware

  • Raspberry Pi 3+
  • Raspberry Pi Camera NoIR
  • IR Illuminator light

Server Setup

Copy over the server/ directory to your choice of hosting service (or home server). Make sure Python 3 and Pipenv are installed. Run pipenv sync to make sure all dependencies are installed. Run pipenv run uvicorn main:app to start.

I recommend using something like Supervisor to manage the server process. See this section for an example.

Create credentials

Run the following command and follow the prompts to create a new username and password that you will use to login:

python cli.py create_user

Next you will need to create an access token for the Raspberry Pi to connect to the server (using an auth token here protects us from letting anything that knows about our server from connecting and sending unwanted data):

python cli.py create_access_token

Copy the access token and save it in a safe place. This is the only time it is shown to you, and it is never stored in plain text in the server data files.

Raspberry Pi Setup

Copy pi/stream.py onto the Raspberry Pi. I'm using Supervisor as the process manager, so the stream client runs automatically, even when the device reboots.

The setup process looks like this:

# Install supervisor
sudo apt-get update
sudo apt-get -y install supervisor

HOSTNAME=<hostname> # enter your server's hostname here
ACCESS_TOKEN=<access_token> # enter your server's access token

# Add the configuration to run the stream demo
echo "\
[program:stream]
command=python3 stream.py --host ${HOSTNAME} --token ${ACCESS_TOKEN}
directory=/home/pi/
user=root
autostart=true
autorestart=true
stopasgroup=true
killasgroup=true" >> stream.conf

sudo mv stream.conf /etc/supervisor/conf.d/

# Tell supervisor to load in new configurations and start your service
sudo supervisorctl update

Creating a self-signed certificate for debugging

If you want to experiment running the server and your Raspberry Pi on your local network, you will need to make sure a proper certificate is set up to allow for the TLS secure socket to function properly.

Create a self-signed certificate for debugging (check out this tutorial for more details):

sudo openssl req \
    -x509 \
    -nodes \
    -days 365 \
    -newkey rsa:2048 \
    -keyout certs/selfsigned.key \
    -out certs/selfsigned.crt
  • When answering the questions, be sure to remember to use the right host name as the FQDN.
  • You may need to enable read permissions. You can use sudo chmod a+r certs/selfsigned.*.

Now you want to add this cert to the Raspberry Pi's ca-certificates list (more details here).

Create a local cert directory:

mkdir /usr/share/ca-certificates/local

Copy selfsigned.crt to this directory on your Raspberry Pi, then reconfigure the ca-certificates package:

dpkg-reconfigure ca-certificates

When prompted, choose the ask option, then select your new certificate by pressing the space key.

Test that it works and then you're ready for local network debugging!

Resources