How to

The following notebooks are illustrated “how-to”, presenting a more advance usage of the methods and algorithms to achieve complex tasks such as cell segmentation or temporal registration.

How-to

• How to organize and reference your data
  • Example
  • Dataset names
  • RAW data
• How-to performs manual 3D registration of intensity images
  • Manual definition of landmarks
• How to performs cell based segmentation with the watershed algorithm
  • Pre-processing
  • Processing
• How to create an intensity image from a segmentation
  • Load the example cell segmented image
  • Hollow out the cells
  • Create a membrane image from the labelled image
  • Make it a bit more realistic
• Manual landmarks definition using intensity projections and altitude maps
  • Use known transformation from create_trsf
  • Surface landmarks GUI:
  • Show associated altitude maps
  • Estimated transformations from landmarks
  • Estimate the deviation from ‘ground truth’ landmarks coordinates:
• Multi-angle fusion on reference image (non-iterative)
  • Get the shared multi-angle intensity images
  • Load the multi-angle images:
  • Down-sample the images to speed-up the computations
  • Intensity projections for multi-angle images
  • Initialisation using manually defined landmarks
  • Fusion on reference image
  • Combine the reference and transformed images with specified fusion method:
  • Centering the mean image in the reference frame:
• How to - Iterative multi-angle images fusion
  • Get the shared multi-angle intensity images
  • Load the multi-angle images:
  • Down-sample the images to speed-up the computations
  • Intensity projections for multi-angle images
  • Load the manual registration landmarks
  • Create manual initial transformations with pointmatching algorithm
  • Iterative fusion
• How to register a temporal sequence of intensity images
  • Introduction
  • Down-sample the images to speed-up the computations
  • Performs sequence rigid registration
• How to perform intensity image registration with manual initialisation.
  • Loading the shared data
  • Displaying the shared data
  • Initial transformation from manually defined landmarks
  • Refining the registration
• Image averaging with VT
  • Load the reference image
  • Create the floating image with a rigid transformation
  • Invert the transformation to create a registered image
  • Create the masks
• VT - How to work with transformations
  • Create transformations
  • Apply transformations
• Usage examples of VT tools
  • Using cellProperties
  • Using vtCellProperties
  • Using handmade test array
• Blockmatching example with vt-python
  • Defines location of shared data:
  • Estimation of rigid transformation
  • Apply transformation to floating image
  • OPTIONAL - TimageTK - Visualization of the rigid registration
  • Estimation of affine transformation
  • Estimation of deformable transformation
• VT-Python - Memory management
  • Load a test image
  • Print a summary of the loaded image:
  • Accessing the reference count of a python object:
  • Check vt.Image object memory address
  • Check the memory address of .to_array() method
  • Check the memory address of .copy_to_array() method
  • Accessing an object size in bytes: