ValueError when trying to reconstruct CT images with ASTRA ToolBox

Question

I'm trying to reconstruct CT images stack, the number of images is 3500, the dimensions of each image is 2560*2160. I'm getting this error: ValueError: The dimensions of the data do not match those specified in the geometry: (3500, 2160, 2560) != (2160, 2560) Thanks for help!

import numpy as np
import astra
import tifffile

Load the TIFF file containing the CT data

data = tifffile.imread(r"E:\Norm.tif")

Define the reconstruction geometry

vol_geom = astra.create_vol_geom(data.shape[1], data.shape[2], data.shape[0])

Define the projection geometry

angles = np.linspace(0, 2*np.pi, data.shape[0], False)
proj_geom = astra.create_proj_geom('parallel', 1.0, data.shape[2], angles)

Create a projector object

proj_id = astra.create_projector('linear', proj_geom, vol_geom)

Create a sinogram from the CT data

sinogram_id, sinogram = astra.create_sino(data, proj_id)

Create a reconstruction object

recon_id = astra.data3d.create('-vol', vol_geom)

Set the parameters for the reconstruction algorithm

cfg = astra.astra_dict('SIRT3D_CUDA')
cfg['ProjectionDataId'] = sinogram_id
cfg['ReconstructionDataId'] = recon_id
cfg['option'] = {'MinConstraint': 0, 'MaxConstraint': 0.02}

Run the reconstruction algorithm

alg_id = astra.algorithm.create(cfg)
astra.algorithm.run(alg_id)

Get the reconstructed volume

recon = astra.data3d.get(recon_id)

Save the reconstructed volume as a TIFF file

tifffile.imwrite('reconstructed_volume.tif', recon)

Answer 1

For 3D imaging, you need a 3D projector geometry: astra.create_projector('parallel3d',...