That projects represents a python-netsocket (server) to manage a Raspberry Pi HQ-Camera based on libcamera (picamera2) as scientific imaging sensor for the measurement of field emission electron sources.
The scripts are developed and tested on a Raspberry Pi 4 Model B (8GB).
1.) Operating System
Install Raspbian OS 11 x64 (no desktop) or higher on a SD-Card. You can use the Raspberry Pi Imager (easiest way) or download the sd-card-image and use your favourite disk-imager software.
2.) Picamera2
As we are using the system headless, the GUI-package installations are skipped.
You can install it directly via apt or also pip. When using pip, some dependencies need to be installed first:
sudo apt install -y python3-libcamera python3-kms++
sudo apt install -y python3-prctl libatlas-base-dev ffmpeg python3-pip
pip3 install numpy --upgrade (on Pi5 try ``sudo apt install python3-numpy``)
Afterwards you can run
sudo apt install -y python3-picamera2 (apt-install)
or
pip3 install picamera2 (pip)
to install the python-class for libcamera.
3.) Additional Software
Some features of picamera2 can use parts of the following packages:
sudo apt install -y python3-opencv
sudo apt install -y opencv-data
pip3 install tflite-runtime (if faced problems, please visit https://stackoverflow.com/questions/74968761/pip3-cant-download-the-latest-tflite-runtime)
sudo apt install -y ffmpeg
1.) Disable legacy camera stack by running sudo raspi-config, enter interface options and make sure, that the legacy camera stack is disabled.
2.) Clone the repository content directly in to the pis home-folder (rPiHQCamServer2.py in /home/pi)
3.) You can adjust some startup-options by opening and modify the rPiHQCamServer2.py, e.g.: srvr_ClipWinBayer: Adjusts the image size in bayer-space before any post-processing or saving.
4.) Run the script by hand (or start it automatically via a ssh-connection)
5.) Use your measurement-programm to connect to the server via the pis IP or DNS and port.
6.) Query commands (Details in the code-documentation of the functions):
- CAM:CONF:SS Adjusts the ShutterSpeed/Exposuretime
- CAM:CONF:FR Adjusts the FrameRate
- CAM:CONF:AG Adjusts the AnalogGain
- CAM:CONF:AWB Adjusts the AutoWhiteBalance
- CAM:CONF:SCLCRP ScalerCrop-functionality not used, because done by SRV:BCLP
- CAP:SEQFET FETches a SEQuence of images; Timeout not used at the moment
- SRV:ARCHV Archvies the given folder to a tar or tar.gz
- SRV:IMG:BCLP Sets the image size. Clipping is done in bayer-space directly after receiving from camera.
- SRV:IMG:DBAY (Post-processing) Sets if the pi is debayering the images before saving
- SRV:IMG:SHRNK (Post-processing) Sets the pi to do pixel-binning
- IDN? Grabs information from the pi (can be used for connection test)
- SRV:ECHO Echoes the given message (an be used for connection test)
- SRV:PATH:RDDIR? Returns the path where the RamDisk is mounted.
- SRV:PATH:SDDIR? Returns the path where SD-Card images captured (is just a shortcut, which is changed by hand in script-code (imFolderPath))
- SRV:PATH:IMDIR? Returns the path where the images stored.
- SRV:CLOSE Closes the connection and shuts down the pycam-server (not the pi)
7.) Images can be downloaded via a SCP-connection from your measurement-program asynchrone from the pi.
This is also hardly recommended, as the images can become huge and cause may an out of RAM/Diskspace exception which crashes the script.
If you don't want to make the effort to program a downloader, you can also try to use a bigger SD-Card and change the image-folderpath from:
imFolderPath = join(mntPnt_RAMDisk, "Captures") # Path to the RAMDISK + Subfolder
to
imFolderPath = join(SDCardPath, "Captures") # Path to the SD-Card + Subfolder
Written by haum