Python library for calibration and control of deformable mirrors (DM). This library implements the methods described in detail in this tutorial from aomicroscopy.org.
dmlib
is meant to be used as an external library in a Python application that
needs to control DMs. dmlib
is not bound to particular hardware, and can be
used with any kind of DM. To control actual hardware you can either use your
own manufacturer-provided library or one of the existing wrappers found in the
subpackage devwraps.
There are two possible use cases:
- Full calibration and control using
dmlib
. You can usedmlib
's GUIs and embed them into your Python application. In this case your DM and scientific camera must be directly supported via the hardware wrappers subpackage devwraps. A guide explaining how to use the GUIs is found here. - Calibration with
dmlib
and control using your own code. If your DM is directly supported bydmlib
, then you can usedmlib
's calibration GUI to collect the calibration data and to generate a calibration file. Subsequently you can load the control matrix from the calibration file and use it in your own code. Instead, if your hardware is not directly supported bydmlib
, you can adjust the data collection scriptdata_collection.py
in theexamples
folder and collect the calibration data with this script. Afterwards, you can generate the calibration as shown in theexamples/calibration.py
script.
Hardware devices such as DMs and scientific cameras are only supported in
Windows via the devwraps
subpackage. You can still install dmlib
under Linux and run the GUIs, but
only in simulation mode (using dummy virtual devices).
To install dmlib
in Windows, follow the steps below.
- You should first install the following software requirements:
- Anaconda for Python 3. This includes Python as well as some necessary scientific libraries.
- Build Tools for Visual Studio. Note that this is not Visual Studio itself, but just the command-line interface Build Tools for Visual Studio 2019. You can find that under Tools for Visual Studio.
- Git. This is necessary for the automatic version numbering of this package. Also, make sure you choose Git from the command line and also 3rd-party software in Adjusting your PATH environment.
- Clone this repository using Git. From any folder in File Explorer,
right-click and hit Git Bash here. Paste
git clone --recurse-submodules https://github.com/jacopoantonello/dmlib
and hit enter. Do not use GitHub's Download ZIP button above, as the installation script will not work in that case. - Finally, double-click on
install.bat
. This script installsdmlib
and its two subpackageszernike
anddevwraps
.
The calibration GUI dmlib.gui
can be used to interferometrically calibrate
and test a DM. You can start the GUI using a BAT script as outlined in the
guide.
Alternatively, open an Anaconda Prompt and query the available drivers with
python -m dmlib.gui --help
. Select the appropriate drivers flags and run the
GUI again without the --help
flag.
For example
python -m dmlib.gui --dm-driver bmc --dm-name DM_SERIAL_NUMBER --cam-driver
thorcam --cam-name CAM_SERIAL_NUMBER
would run dmlib.gui
using the Boston Multi-DM with serial number
DM_SERIAL_NUMBER
and the Thorlabs camera with serial number
CAM_SERIAL_NUMBER
.
To run the calibration GUI in simulation mode without any hardware, use
python -m dmlib.gui --dm-driver sim --cam-driver sim
For convenience, you can create a BAT
file with the correct parameters by
modifying the examples/run_calibration_gui.bat
.
The control GUI dmlib.zpanel
can load a DM calibration file and control the
DM in open-loop using Zernike modes. You can start the GUI using a BAT script
as outlined in the
guide.
Alternatively, open an Anaconda Prompt and use the -m dmlib.zpanel
flag to
start this GUI. You can query the GUIs flags with --help
as seen above.
[1] J. Antonello, J. Wang, C. He, M. Phillips, and M. Booth, "Interferometric calibration of a deformable mirror," 10.5281/zenodo.3714951.