unravel.image_io.io_nii module#
Use io_nii from UNRAVEL to convert the data type of a .nii.gz image and optionally scale the data.
- Inputs:
.nii.gz image
- Outputs:
.nii.gz image with the specified data type and scaling.
- Possible numpy data types:
Unsigned Integer: uint8, uint16, uint32, uint64
Signed Integer: int8, int16, int32, int64
Floating Point: float32, float64
Note
With –scale, the min intensity becomes dtype min and max intensity becomes dtype max. Every other intensity is scaled accordingly.
With –binary, the image is binarized (0 or 1).
With –zscore, the range of z-scored data from -3 to 3 is converted to 0 to 255.
With –fixed_scale, the data is scaled using the provided min and max values.
Usage:#
io_nii -i path/img.nii.gz -d float32 [-o path/new_img.nii.gz] [-f min,max | -s | -b | -z] [-v]
- unravel.image_io.io_nii.convert_dtype(ndarray, data_type, scale_mode='none', fixed_scale_range=None, zscore_range=(-3, 3), target_range=(0, 255))[source]#
Convert the data type of an ndarray and optionally scale the data.
Parameters: - ndarray: Input ndarray. - data_type: Target data type. - scale_mode: ‘none’, ‘standard’, or ‘zscore’. Determines the scaling approach. - fixed_scale_range: Tuple indicating the fixed range for scaling if scale_mode is ‘fixed’. - zscore_range: Tuple indicating the z-score range for scaling if scale_mode is ‘zscore’. - target_range: Tuple indicating the target range for the data type conversion.
Returns: - Converted ndarray with the specified data type and scaling.