This codec is designed for compressing movies with Poisson noise, which are produced by photon-limited modalities such multiphoton microscopy, radiography, and astronomy.
The codec assumes that the video is linearly encoded with a potential offset (dark_signal) and that the photon_sensitivity (the average increase in intensity per photon) is known or can be accurately estimated from the data.
The codec re-quantizes the grayscale efficiently with a square-root-like transformation to equalize the noise variance across the grayscale levels: the Anscombe Transform.
This results in a smaller number of unique grayscale levels and improvements in the compressibility of the data with a tunable trade-off (beta) for signal accuracy.
To use the codec, one must supply two pieces of information: dark_signal (the input value corresponding to the absence of light) and photon_sensitivity (levels/photon). We provide two alternative routines to extract those numbers directly from signal statistics. Alternatively, they can be directly measured at the moment of data acquisition. Those calibration routines are provided in the src/poisson_numcodecs/calibrate.py file.
The codec is used in Zarr as a filter prior to compression.
Zarr.
Install via pip:
pip install poisson-numcodecs
conda create -n poisson_numcodecs python=3.xx
conda activate poisson_numcodecs
git clone https://github.com/AllenNeuralDynamics/poisson-numcodecs.git
cd poisson-numcodecs
pip install -r requirements.txt
pip install -e .
Make sure everything works:
pip install pytest
pytest tests/
A complete example with sequential calibration and look-up compression is provided in examples/raster_calibration_and_compression.ipynb A complete example with raster calibration and compression is provided in examples/sequential_calibration_and_lookup_compression.ipynb
This fork provides additional scripts for specific tasks. After installation, you can use the following commands:
To create a sample figure displaying the photon-transfer curve, coefficient of variation, and quantum flux, run:
python examples/make_figure.py input_image.tif output_path/ --title "Title of the figure"Replace input_image.tif with your input image file and output_path/ with the desired directory for the output. The optional --title argument allows you to specify a custom title for the figure.
To rescale your movie to photon/pixels/frame, use:
python examples/convert_movie.py input_movie.tif output_movie.tifReplace input_movie.tif with the input movie file and output_movie.tif with the desired output file name.