Add prototype for OME-ZARR export

CHANGELOG.md

@@ -1,3 +1,58 @@
+## Release September 2025 [1.11.1]
+:rocket: Overhaul of internal bootlnecks that hampered performance on some systems and caused GUI freezing / high CPU loads.
+### Bugfixes :bug: 
+- fixed excessive communication between threads causing high CPU load on some systems.
+- moved `serial_worker` into `core` thread to avoid conflicts of relative motion operations with GUI thread.
+- added CPU core identifiers for different operations in `debug` mode or logging, to pinpoint performance issues.
+- fixed autofocus (AF) function in the GUI.
+- returned `'camera_display_live_subsampling': 2, `,  `'camera_display_acquisition_subsampling': 2,` and `'camera_display_temporal_subsampling': 2,` into the config file, to reduce camera display load on older computers.
+- fixed light-sheet markers and box ROI markers in the Camera window, which were not displayed correctly after zoom change.
+
+## Release March 2025 [1.11.0].
+### Performance :rocket:
+- implementation of time lapse function via script. See example script at `mesoSPIM/scripts/timelapse.py` for details.
+- images can now be displayed at full resolution during either live mode or acquisition, with no performance penalty. Deprecated config parameters:
+``` python
+'camera_display_live_subsampling': 2,
+'camera_display_acquisition_subsampling': 2, 
+```
+to enjoy full-resolution image in real time. The corresponding GUI elements for controlling the downsampling were removed.
+
+### Bugfixes :bug: 
+- duplication of sub-stacks while writing large H5 files (>0.5 TB) and freezing at the end of acquisition. Fixed by replacing signal/slot mechanism for CameraWindow image update to deque mechanism. Boosted performance and stability.
+- `MAX` projection is always on by default, if the dataset is asvaed in TIFF or RAW format. No need to check the box in the Image Processing Wizard. This generates maximum intensity projection as TIFF file for each stack in the acquisition list.
+- explicit initialization of the first raw in Acquisition Manager, to avoid undefined GUI widget states.
+
+## Release December 2024 [1.10.2].
+### Performance :rocket:
+- migration to Python 3.12 for better performance and compatibility with the latest libraries.
+- faster loading and response time of the GUI
+- subsampling of the camera image for less frequent rendering in the GUI (less CPU load) during acquisition, controlled by `'camera_display_temporal_subsampling': 2,` parameter in the config file. Value `1` means no subsampling (show every image during acquisition).
+- change recommended package manager to `mamba` to avoid Anaconda license issues.
+- change the default timer for stage position update from 50 ms to 100 ms, to reduce communication and logging overhead. ASI stages updated every 200 ms, PI stages every 100 ms.
+- zoom dictionary can include multi-word names like `5x Mitutoyo` instead of `5x` for better UI and meta-infomation.
+- spinboxes are limited in scroll speed to allow hardware catch up with UI in the interactive mode
+- #82: uploading UCL-Bechtop config file `config/examples/config_benchtop_UCL_5laser.py` by @TchLenn and parsing of double-digit strings like `ao21:22` in config file.
+- remove `Log` tab from the GUI, as it was not used by the users and was causing unnecessary overhead, esp. in DEBUG mode.
+
+### User Interface :lollipop:
+- ETL config files (.csv) are automatically checked and updated with the `laser`-`zoom` combination selected in GUI, to avoid errors in the acquisition. User can even start with an empty ETL config file, it will be auto-filled with the `laser`-`zoom` combinations on the go.
+- more tooltips added to the GUI elements for better user experience.
+
+### :wrench: system changes
+- pip depenencies are frozen for staibilty
+- `mesoSPIM_State` is only nominally a singleton, but actually inherited by classes from their parent class to ensure unique state and thread safety.
+- no signal from `mesoSPIM_State` every time the state is updated. The GUI is updated from state by a separate signal/slot from member classes, on demand.
+- bugfix: serial communication with ASI stages is now thread-safe, with no conflicts between GUI (Main) and Core threads.
+- bugfix: parsing config file with either `'laser_task_line' :  'PXI6733/ao0:3'` or `'/PXI6733/ao0:3'` notation (Alan Watson).
+
+## Release candidate August 2024 [1.10.0]. Cancelled due to performance issues on some setups.
+### Performance :rocket:
+- improved acquisition speed from 2.5 to 5 FPS, by moving image writing to a separate thread and using `collections.deque` mechanism for frame sharing between the camera and image writing threads, instead of less-performant `QThread` signal/slot mechanism.
+- best writing speed and lowest CPU overhead is achieved with NVMe SSD disks, recommended.
+- the option `buffering = {...}` in config file is deprecated from v.1.10.0 and will be ignored, due to improved program performance.
+- dependencies are updated to the latest versions: `numpy`, `scipy`, `tifffile`, etc.
+
 ## Release July 2024 [1.9.0]
 ### User Interface :lollipop:
 - :gem:  "Auto L/R illumination" button in the Acquisition manager to select tile illumination based on its x-position.

README.md

@@ -1,23 +1,26 @@
 [![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.6109315.svg)](https://doi.org/10.5281/zenodo.6109315)
-[![Python 3.7](https://img.shields.io/badge/python-3.7-blue.svg)](https://www.python.org/downloads/release/python-370/)
+[![Python 3.12](https://img.shields.io/badge/python-3.12-blue.svg)](https://www.python.org/downloads/release/python-312/)
 [![License: GPL v3](https://img.shields.io/badge/License-GPLv3-blue.svg)](https://www.gnu.org/licenses/gpl-3.0)
+[![Static Badge](https://img.shields.io/badge/user_forum-image.sc-blue)](https://forum.image.sc/tag/mesospim)
 
-# mesoSPIM-control
-Image acquisition software for [mesoSPIM](http://mesospim.org/) light-sheet microscopes. 
-A mesoSPIM (mesoscale selective plane illumination microscope) is optimized for fast imaging of large (many cm³) cleared tissue samples at near-isotropic resolution. 
-Currently, more than 20 mesoSPIM setups are operational [around the world](http://mesospim.org/setups/).
 
-Parts lists, drawings, and instructions for building a mesoSPIM can be found in the [mesoSPIM wiki](https://github.com/mesoSPIM/mesoSPIM-hardware-documentation).
+# mesoSPIM-control
+Image acquisition software for [mesoSPIM](http://mesospim.org/) light-sheet microscopes. Compatible with all official versions of mesoSPIM hardware (Benchtop, v4, v5), and multple alternative configurations.
 
 ## Overview
-The mesoSPIM is a versatile light-sheet microscope for imaging
-cleared tissue samples. It is compatible with all major clearing approaches and optimized for quickly creating large-field-of-view overview datasets.
+The mesoSPIM (mesoscale selective plane illumination microscope) is a family of versatile open-source microscopes optimized for fast imaging of large (many cm³) cleared tissue samples at near-isotropic resolution. 
+Currently, more than 30 mesoSPIM setups are operational [around the world](http://mesospim.org/setups/).
+Parts lists, drawings, and instructions for building a mesoSPIM can be found in the wiki pages:
+- [Benchtop mesoSPIM](https://github.com/mesoSPIM/benchtop-hardware)
+- [mesoSPIM v4-5](https://github.com/mesoSPIM/mesoSPIM-hardware-documentation)
+
+The `mesoSPIM-control` is python-based acquisition software with user-friendly GUI, based on PyQt5 framework.
 
 ## Installation
 
 ### Prerequisites
-* Windows 7 or Windows 10, 64-bit
-* Python >=3.7, we recommend [Anaconda](https://www.anaconda.com/download/)
+* Windows >=7, 64-bit
+* Python >=3.12, we recommend [Miniforge](https://github.com/conda-forge/miniforge) and `mamba` package manager that comes with it. The Anaconda distribution is not recommended since the change of its terms of service in 2020. If you have existing Anaconda installation, you can still use `conda` instead of `mamba` in the instructsions below.
 
 ### Device drivers
 #### National Instruments DAQ devices
@@ -36,32 +39,32 @@ When installing the MS Visual C++ tools, make sure to check [C++ build tools](ht
 * ASI stages: [ASI Tiger drivers](http://www.asiimaging.com/support/downloads/tiger-controller-console/). 
 If using USB connection, check ASI instructions on [USB support](http://www.asiimaging.com/support/downloads/usb-support-on-ms-2000-wk-controllers/)
 
-## Installation steps 
+### Installation steps 
 1. Clone this repository either by via GitHub Desktop (recommended) or by downloading and unpacking the ZIP file into folder `C:/Users/Public/mesoSPIM-control`
 
 ![image](https://user-images.githubusercontent.com/10835134/198991579-df1e5acc-d246-425b-a345-03ba93a1f0bb.png)
 
-2. Open Anaconda prompt, create and activate a new environment `mesoSPIM-py37`:
+2. Open Miniforge prompt, create and activate a new environment `mesoSPIM-py312`:
 ```
-conda create -p C:/Users/Public/conda/envs/mesoSPIM-py37 python=3.7
-conda activate C:/Users/Public/conda/envs/mesoSPIM-py37
+mamba create -p C:/Users/Public/mamba/envs/mesoSPIM-py312 python=3.12
+mamba activate C:/Users/Public/mamba/envs/mesoSPIM-py312
 ```
 3. Install mesoSPIM-specific libraries: 
 ```
 cd C:/Users/Public/mesoSPIM-control
-pip install -r requirements-anaconda.txt
+pip install -r requirements-conda-mamba.txt
 ```
 
 ## Launching
 
-### Anaconda prompt
+### Miniforge prompt
 1. `cd C:/Users/Public/mesoSPIM-control/mesoSPIM`
 2. `python mesoSPIM_Control.py` (with argument `-D` for demo mode)
 
 ### Desktop shortcut (fast launch)
-From Anaconda prompt, type `where conda`, and enter the result (e.g. `C:\Users\Nikita\anaconda3\Scripts\activate.bat`) into line 10 of `mesoSPIM.bat` file:
+From Miniforge prompt, type `where mamba`, and enter the result (e.g. `C:\Users\Nikita\miniforge3\Scripts\activate.bat`) into line 10 of `mesoSPIM.bat` file:
 ```
-"%windir%\System32\cmd.exe" /k ""C:\Users\Nikita\anaconda3\Scripts\activate.bat" "C:\Users\Public\conda\envs\mesoSPIM-py37" && python "mesoSPIM_Control.py""
+"%windir%\System32\cmd.exe" /k ""C:\Users\Nikita\miniforge3\Scripts\activate.bat" "C:\Users\Public\conda\envs\mesoSPIM-py312" && python "mesoSPIM_Control.py""
 ```
 Save changes and double-click the `mesoSPIM.bat` file - this should launch the control software. If this does ot happen, check the Anaconda path. Once this works, create a shortcut and place it on your desktop for quick launching.  
 
@@ -75,18 +78,20 @@ If you have multiple config files you will be prompted to choose one that corres
 Once your hardware is connected and turned on, change the `Demo` devices to hardware-specific names, set their parameters, and test each device.
 See [Wiki](https://github.com/mesoSPIM/mesoSPIM-hardware-documentation/wiki/mesoSPIM_configuration_file) for details.
 
-
-
 ## Updating existing installation
-To ensure safe transition to a new version, we recommend fresh installation of each new version into a separate folder (e.g. `mesoSPIM-control-Nov2022`) using the steps above. In order to unlock all new features, please review and add new sections from the [demo config file](/mesoSPIM/config/demo_config.py) to your old configuration file.
+To ensure safe transition to a new version, we recommend fresh installation of each new version into a separate folder (e.g. `mesoSPIM-control-Jan2025`) using the steps above. In order to unlock all new features, please review and add new sections from the [demo config file](/mesoSPIM/config/demo_config.py) to your old configuration file.
+
+## Troubleshooting
+Some errors during software startup and acquisition can be displayed in the mamba/conda terminal, while a highly detailed log will be recorded in the time-stamped file in `mesoSPIM/log` folder. This log file is individual for each session, named by date and time of software start-up, eg `mesoSPIM/log/20241210-154845.log`.
 
-## Documentation for users
+## Further resources
 For instructions on how to use mesoSPIM-control, please check out the documentation:
-* [PPT](https://github.com/mesoSPIM/mesoSPIM-powerpoint-documentation), 
+* ZMB (Univerity of Zurich) guides on [mesoSPIM start-up, setting up, and acquisition](https://zmb.dozuki.com/c/Lightsheet_microscopy#Section_MesoSPIM)
+* [PPT](https://github.com/mesoSPIM/mesoSPIM-powerpoint-documentation) (outdated version), 
 * youtube [channel](https://www.youtube.com/c/mesoSPIM), 
 * subscribe to our [mailing list](http://eepurl.com/hPBRhj).
+* join the discussion at the [image.cs user forum](https://forum.image.sc/tag/mesospim) (since August 2025)
 
-If you have questions, contact the current core developer [Nikita Vladimirov](mailto:vladimirov@hifo.uzh.ch).
 
 ## How to cite this software
 Fabian F. Voigt, Nikita Vladimirov, Christian Schulze, Rob Campbell, & Fritjof Helmchen. (2022). MesoSPIM control: An open-source acquisition software for light-sheet microscopy written in Python and Qt. Zenodo. https://doi.org/10.5281/zenodo.6109315