Table of Contents

• Overview
• Main Features
• Requirements
• Installation
• Usage
  • Camera Control
  • Image and Video Capture
  • Motion Detection
  • Timelapse Feature
  • Camera Settings
  • Configuration and Status
  • Filename Creation
  • Stopping the Program
• Contributing
  • Issues
  • Code Changes
  • Testing Changes
• Acknowledgements

Overview

RasPyCam is a python-based multi-stream camera system for the Raspberry Pi. This application has been designed as a replacement for the RasPiCam application developed by Silvan Melchior, Robert Tidey and others which is no longer maintained and has compatibility issues with the latest versions of the Raspberry Pi.

There are two ways of using RasPyCam:

1. As a standalone program: The program can be run on a Raspberry Pi with a camera module connected to it. The program will respond to manual command calls and perform the necessary actions. This method will use the source code provided in this repository.
2. As a backend service: The program can be run on a Raspberry Pi with a camera module connected to it alongside the RPi Cam Web Interface system. The program will respond to the commands sent by the frontend and perform the necessary actions. This method will use the executable file created from the source code provided in this repository. As of version 1, the executable file is shipped with the RPi Cam Web Interface system. Future updates to this repository will update the executable file in the frontend system.

Main Features

• Preview: Retrieve the camera feed in real time.
• Image Capture: Capture images from the connected camera.
• Video Recording: Record videos from the connected camera.
• Motion Detection: Detect motion in the camera feed.
• Timelapse Image Capture: Capture periodic images from the connected camera.
• Multi-Stream Support: Stream multiple camera feeds simultaneously from the connected camera.
• Web Interface Support: Interact with the program using the RPi Cam Web Interface system.

Requirements

Depending on how you run the program, you may need to install some dependencies, including PiCamera 2, OpenCV, and Pillow.

If you would like to run the source code, you can install these libraries using the following commands:

pip install picamera2 opencv-python Pillow

Installation

1. Clone the repository to your Raspberry Pi:

git clone https://github.com/windermere-technology/raspycam.git

2. Run the installation script to install the program and its dependencies:

[!NOTE] Depending on your Pi's configuration, you may need to run the program with sudo privileges. This is because the program requires read/write access to the tmp and var directories which are restricted to root access.

sudo ./install.sh

You will be prompted if you want the frontend (RPi Cam Web Interface) to be install alongside the program. Select y if you want to install the frontend, otherwise select n.

After the install script finished, you can run the program standalone by running the following command, or it will run automatically with the frontend:

raspimjpeg [--config /path/to/config1 /path/to/config2 ...]

path/to/config is the path to the configuration file you want to use. If you don't specify a configuration file, the program will use the default configuration file provided in the repository. Provide 2 config will apply the settings to 2 cameras.

Usage

RasPyCam is a continuously running program that observes the commands sent to the named pipe (by default this is /var/FIFO).

To send commands to the program, you can use the following commands:

echo '{command} {parameter}' > /var/FIFO

[!NOTE] While the front end is running, the RasPyCam program will be running under www-data user. This means that the FIFO file will be located in /var/www/html/FIFO instead of /var/FIFO. And you will need to send commands as www-data user.

sudo su -c "echo '{command} {parameter}' >> /var/www/html/FIFO" www-data

[!NOTE] Angled brackets refer to \<optional parameters>.

[!NOTE] Square brackets are used to send commands to multiple cameras. Command and sets of parameters are separated by commas. Forward slashes escape commas. Leading and trailing whitespaces are not stripped from parameters. Not enclosing parameters within brackets applies the parameters to all bracketed commands.

Example 1: [dp,dp] [0,1] = Switches off preview for Camera 0, switches on preview for Camera 1.

Example 2: [im,an] [,Look/, an annotation!] = Takes a still image with Camera 0, sets the annotation for Camera 1 to "Look, an annotation!".

Example 3: [,im] = Camera 0 is not sent any command. Takes an image with Camera 1.

Example 4: [ca,ca] 1 = Starts recording on both Camera 0 and Camera 1.

[!IMPORTANT] The commands ix+ix and im+im can only be used within square brackets, even if only sent to 1 camera.

Currently the program supports the following commands:

Camera Control

Image and Video Capture

Motion Detection

Timelapse Feature

Camera Settings

Configuration and Status

Command names, parameters and paths have been sourced from the RPi Cam Web Interface system to ensure compatibility. In addition to the existing naming scheme used to interpolate values into filenames, the code %I can be used to refer a camera's index number.

Filename Creation

Filenames can be created with standard text alongside the following naming scheme:

[!NOTE] The %a code will read the text from the /dev/shm/mjpeg/user_annotate.txt file by default. This file must be created by the user. You may change this default path in the configuration file by adjusting the user_annotate parameter.

Stopping the Program

To stop the program, you can either send SIGINT or SIGTERM signals to the program. This can be done by either pressing Ctrl+C in the terminal running the program or by using the kill command.

If you've launched the program using the source code, run the following command:

sudo kill -9 $(cat /opt/vc/bin/raspycam/raspy.pid)

Or if you've launched the program using the front-end, navigate to the front-end directory and run the following command, or simply press the stop button in the front-end:

sudo ./stop.sh

Contributing

Contributions to the RasPyCam project are welcome. If you would like to contribute to the project, please follow the steps below:

Issues

If you encounter any issues with the program, you can create a new issue on the Issues page and use one of the provided templates to report the issue.

Code Changes

To make changes to the code, you can follow the steps below:

1. Fork the repository and clone it to your local machine
2. Make your changes, commit and push them to your forked repository
3. Create a pull request back to the main repository (Windermere-Technology/RasPyCam)
4. Wait for the pull request to be reviewed and merged

Testing Changes

To test your changes and generage a coverage report, you can run the following commands on a Raspberry Pi with a camera module connected to it:

sudo apt update
sudo apt install libcap-dev python3-pytest libopencv-dev python3-pytest-cov -y

PYTHONPATH=./app pytest --import-mode=importlib --cov=app --cov-report=term --cov-report=html:coverage_html --cov-config=tests/.coveragerc

This command will run the testing suite and generate a coverage report. You can view this report by opening the coverage_html/index.html file in your browser. If you want to test a specific folder or file, just add the path to the end of the command.

Acknowledgements

The development of this project was inspired by the RasPiCam application developed by Silvan Melchior, Robert Tidey and others.

Initially devleoped as part of a University project overseen by Cian Byrne, the development team consisted of:

• Kyle Graham
• Harry Le (Lê Thành Nhân)
• Chen-Don Loi
• Qiuda (Richard) Song