User manuals MuhRec BatchProcessing
The script shown on the reconstruct from CLI page requires some thinking and coding to put it in order. Batch processing was developed to reduce the need for programming. Instead, you'll need to provide only a file describing the batches to reconstruct.
You need the following to run the batch scripts:
- Python 3.5 or higher
- The following scripts
batch_runner.py-
goldenbatch.py(optional)
The batches to process are described in a json formatted file with the following format
[{
"name" : "Reconstruction 1",
"config" : "path/to/template/reconconfig.xml",
"arguments" : {
"parameter 1" : "value 1",
"parameter 2": "value 2",
"parameter 3": "1 2 3 4 5"
}
}
]As you can see a list of entries and it is important to use the opening and closing brackets [...] and each batch block must be separated by commas ,.
The name element is only a label to describe the current batch. The value can be anything.
The config element is the path to the reconstruction configuration file you want to base the reconstruction upon. This file is generated by muhrec every time you reconstruct some data and stored in \your\home\folder\.imagingtools\CurrentRecon.xml. It is also saved in the same folder as your reconstructed slices when you save in the muhrec GUI.
The arguments are the parameters you want to change for the reconstruction batch. The arguments override the parameters in the reconconfig.xml file. It is often the projection data path, destination path, the first and last projection index, etc.
Currently, it is only possible to manipulate the parameters in the <system>, <projections>, and <matrix> blocks of the ReconConfig.xml file. The <processing> and <backprojector> blocks can't be manipulated.
There is a list of arguments on the CLI operation manual page.
[{
"name" : "recon1",
"config" : "path/to/template/reconconfig.xml",
"arguments" : {
"projections:filemask" : "path/to/data/image_####.tif",
"matrix:path": "path/to/matrix/destination",
"matrix:matrixname": "slice_####.tif",
"projections:firstindex": 1,
"projections:lastindex" : 100,
"projections:direction" : "CW",
"projections:scanarc0" : 0.0,
"projections:scanarc1" : 180.0
}
},
{
"name" : "recon2",
"config" : "path/to/template/reconconfig.xml",
"arguments" : {
"projections:filemask" : "path/to/data/image_####.tif",
"matrix:path": "path/to/matrix/destination",
"matrix:matrixname": "slice_####.tif",
"projections:firstindex": 101,
"projections:lastindex" : 200,
"projections:direction" : "CW",
"projections:scanarc0" : 180.0,
"projections:scanarc1" : 360.0
}
}
]The batch_runner script runs a set of reconstruction batches using MuhRec. The batches are specified in a JSON file using the format described above.
To run the batch_runner, you need a terminal window with an available Python installation.
python batch_runner.py [-h] [-b BATCHFILE] [-m MUHRECPATH]| Short | Long | Description |
|---|---|---|
| -h | --help | Show a help message and exit |
| -b BATCHFILE | --batchfile BATCHFILE | A JSON file with batch descriptions. |
| -m MUHRECPATH | --muhrecpath MUHRECPATH | Path to the muhrec executable. |
Prepares a set of reconstruction batches for the batch_runner. The batches are created from a longer series of projections like a golden ratio or a multiple-turn scan. The script produces batches with a given number of projections. This script reduces the risk of typing error while creating the batch file for series.
python goldenbatch.py [-h] [-t TEMPLATE] [-s STEP] [-f FIRST] [-l LAST] [-o OUTPUT]| Short | Long | Description |
|---|---|---|
| -h | --help | Show a help message and exit |
| -t TEMPLATE | --template TEMPLATE | A json file with batch descriptions. The first batch will be used as template for the generated sequence |
| -s STEP | --step STEP | Number of projections to reconstruct in a batch. |
| -f FIRST | --first FIRST | The index of the first projection. |
| -l LAST | --last LAST | The index of the last projection. |
| -o OUTPUT | --output OUTPUT | The name of the produced batch file. |
