Surveillance system based on Raspberry Pi.
Pisurvl is a surveillance solution with motion detection and alert notifications. This repository contains the server part of the solution. It is implemented in Python and it uses OpenCV to access the camera of the Raspberry Pi. Despite being implemented on top of a Raspberry Pi, the solution should work on any system having a camera that is supported by OpenCV.
Currently there is no easy way to install OpenCV 3 on Linux. The following commands install all the necessary dependencies to build OpenCV 3:
sudo apt install build-essential cmake git pkg-config
sudo apt install libgtk2.0-dev libavcodec-dev libavformat-dev libswscale-dev libv4l-dev
sudo apt install libjpeg-dev libpng-dev libtiff-dev libjasper-dev libdc1394-22-dev
sudo apt install libatlas-base-dev gfortran liblapacke-dev
sudo apt install python3 python3-dev python3-setuptools
sudo easy_install3 pip
The following command installs the latest version of Numpy. The installation may take a while to complete.
sudo pip3 install numpy
The following commands download and compile OpenCV 3:
mkdir build-opencv && cd build-opencv
git clone https://github.com/opencv/opencv.git
git clone https://github.com/opencv/opencv_contrib.git
cd opencv_contrib && git checkout 3.2.0 && cd ..
cd opencv && git checkout 3.2.0
mkdir build && cd build
cmake -D CMAKE_BUILD_TYPE=Release \
-D CMAKE_INSTALL_PREFIX=/usr/local \
-D OPENCV_EXTRA_MODULES_PATH=`pwd`/../../opencv_contrib/modules \
..
make -j4
Then install OpenCV:
sudo make install
sudo pip3 install --upgrade pisurvl
git clone https://github.com/betabandido/pisurvl.git
sudo cp pisurvl/init/pisurvl.service /etc/systemd/system
sudo systemctl enable pisurvl.service
Create a settings file based on the provided template:
mkdir -p ~/.pisurvl
cp pisurvl/template-settings.yaml ~/.pisurvl/settings.yaml
Fill in the fields from
and to
in the notifications
section.
Use Google API Manager to generate some client credentials and download the client_secrets.json
file. Then move
the file into the PiSurvl configuration folder:
mv client_secrets.json ~/.pisurvl
Execute pisurvl-setup --noauth_local_webserver
and follow the instructions on the screen to authorize PiSurvl to
access Google Drive.
Use raspi-config
to enable the camera interface. Then, after rebooting, load the Video4Linux driver by running:
sudo modprobe bcm2835-v4l2
Add this module to /etc/modules
to automatically load the module whenever the Raspberry Pi boots.
Use systemctl
to start the service:
sudo systemctl start pisurvl.service
To upload a new release to PyPI follow the next steps:
- Bump version in
setup.py
- Run
python3 setup.py sdist upload