segview

A pure Python module for visualising 3D segmentation results


License
GPL-3.0
Install
pip install segview==0.3.0

Documentation

SegView

visualise the 3D segmentation result

Install

  • The most convenient way would be: pip3 install segview
  • You can also Include the file segview.py in your project directory

(SegView only support Python 3.5+, because it requires PyQt5)

Use SegView

segview.render_label(label, metadata, alpha=1) # see the 3D model of labels

segview.annotate_label(image, label, axis=-1)  # see the 2D slice with labels along different axes

segview.render_image(image, metadata)  # see the 3D render of an image

segview.render_image(image, metadata, feature)  # see the 3D image with features

segview.annotate_feature(image, label)  # see 2D slice with features
  • label is a 3D numpy array
    • Usually it is the result of image segmentation, having the same structure
    • Value 0 corresponds to the background
    • Its shape is (x, y, z).
  • feature is a 2D numpy array
    • Usually it is the result of intensity maxima locating
    • It is 3D positions, [(x1, y1, z1), (x2, y2, z2), ...]
    • Its shape is (feature_number, 3)
  • metadata is a dictionary containing the voxel size
    • It is only used in 3D visualisation, as many z-stack images have lower resolutions along z-axis
    • {'voxel_size_x': 1, 'voxel_size_y': 1, 'voxel_size_z': 1}
  • alpha adjusts the brightness of the result