lcit / ext3dlbp Goto Github PK

Does the library work only for int32 data type for the 3D array?
I have this function that reads a nifti file and return a niiROI (np.ndarray with float type).

def readNifty(filePath):
    image = sitk.ReadImage(filePath)
    print("Reading Nifty format from {}".format(filePath))
    print("Image size: {}".format(image.GetSize()))
    metadata = Metadata(image.GetOrigin(), image.GetSpacing(), image.GetDirection())

    # Converting from SimpleITK image to Numpy array. But also is changed the coordinate systems
    # from the image which use (x,y,z) to the array using (z,y,x).
    volume_zyx = sitk.GetArrayFromImage(image)
    volume_xyz = np.transpose(volume_zyx, (2, 1, 0)) #back to the initial xyz coordinate system
    print("Volume shape: {}".format(volume_xyz.shape))
    print("Minimum value: {}".format(np.min(volume_xyz)))
    print("Maximum value: {}".format(np.max(volume_xyz)))
    return volume_xyz, metadata     # return two items.

When I run this, I have encountered memory error:

(img_3d_NI,img_3d_RD,img_3d_CI) = lbp.convert_3d_image(niiROI) #img3D

MemoryError

Any tips on solving this would be great. Thanks!

Installing OpenCV + Contrib + FFmpeg, such error encountered at the end of make -j8

[ 25%] Built target libprotobuf
CMakeFiles/Makefile2:2452: recipe for target 'modules/core/CMakeFiles/opencv_core.dir/all' failed
make[1]: *** [modules/core/CMakeFiles/opencv_core.dir/all] Error 2
Makefile:160: recipe for target 'all' failed
make: *** [all] Error 2

What might be the problem?

Python 3.7.3 (default, Mar 27 2019, 22:11:17)
[GCC 7.3.0] :: Anaconda, Inc. on linux
Type "help", "copyright", "credits" or "license" for more information.

import ext3DLBPpy
Traceback (most recent call last):
File "", line 1, in
ImportError: /home/kai/masterthesis/ext3DLBP/python_wrapper/build/ext3DLBPpy.so: undefined symbol: _ZTIN5boost6python15instance_holderE

Hello...
In your paper, the joint of NI/RD shows promising results (classification accuracy hits 94.6% for sample rate: 1, 42) and I'd like to reproduce the method for experiments. However the result of that using the given construct_histograms function, was a 3 dimensional array:
joint_2d = np.outer(hist1,hist2) #Joint: NI/RD #(52, 12, 12)

While the concatenations are 2 dimensional (which can be fed to classifiers):

concat_of_ni_rd = np.concatenate([hist1,hist2]) #Concatenation: NI+RD, #(52, 24)
concat_of_ni_rd_ci = np.concatenate([hist1,hist2,hist3]) #Concatenation: NI+RD+CI #(52, 36)

What should it be done to make the joint vectors be ready for training? Thanks!

This example below reads .bmp files. To have it read images from LIDC-IDRI, e.g. LIDC-IDRI-0124_GT1_1.nii.gz, this can be done by modifying this function def from_images_to_3D_array(directory, size)? Thanks.

examples/convert_3d_texture_python/main.py

I was testing the model (lbp = ext3DLBPpy.NI_RD_CI_LBP_P42g_R1(mur, V)) with 50+ nii.gz files. The first 22 files were fine. But when it comes to processing the 23rd file, this error occurred. I have tried different spatial resolution, but the error is the same.

$ python3 test.py 
Reading Nifty format from .../training/LIDC-IDRI-0385_GT1_1.nii.gz
Image size: (15, 18, 15)
Volume shape: (15, 18, 15)
Minimum value: 48.0
Maximum value: 2009.0
python3: malloc.c:4023: _int_malloc: Assertion `(unsigned long) (size) >= (unsigned long) (nb)' failed.
Aborted (core dumped)

I have no idea what this error suggests. What might be the problem, do you think? Thanks.