Jefferis Lab DokuWiki warping_manual

warping_manual:all_affine_parameters

Anonymous (anonymous@undisclosed.example.com) — 2013-12-10T11:59:45+00:00

registration: Voxel-based affine image registration. Syntax: registration [options] reference.study model.study or registration [options] input.list where [options] is one or more from: Flags: --copyright Print copyright message. --copyright-vtk CopyrightVTK -v, --verbose Output diagnostic information (default) -q, --quiet Suppress all diagnostic output -i, --initxlate Initially align centers of volumes -n, --noswap Prevent automatic swapping of model and reference Input …

warping_manual:all_warp_parameters

Anonymous (anonymous@undisclosed.example.com) — 2013-12-10T11:59:45+00:00

warp: Voxel-based affine image registration. (see also affine registration tool) Syntax: warp [options] floating.study reference.study or warp [options] input.list where [options] is one or more from: Flags: --copyright Print copyright message. --copyright-vtk CopyrightVTK -v, --verbose Output diagnostic information (default) -q, --quiet Suppress all diagnostic output -S, --switch Exchange floating and reference image. --exchange Force exchange of floating and reference…

warping_manual:bounding_boxes_and_image_grids

Anonymous (anonymous@undisclosed.example.com) — 2013-12-10T11:59:45+00:00

This is one of the most vexing issues in dealing with image data. When dealing with 2D (or higher dimensional) image data, we need to specify the transformation between some kind of world coordinates and the image voxel data. The basic stuff that we need to know is:

warping_manual:bridging_different_template_brains

Anonymous (anonymous@undisclosed.example.com) — 2013-12-10T11:59:45+00:00

Preview Imagine you have two sets of experiments with two different template brains. You have two large data sets that you would like to compare, but they are in separate frames of reference. You therefore need to compute the transformation that maps one template onto the other. For the purposes of this comparison, you will need to decide which of the two templates is going to be the master template. This is particularly important because the warping registrations are not simply invertible …

warping_manual:calculating_an_initial_registration_from_landmarks

Anonymous (anonymous@undisclosed.example.com) — 2014-06-10T11:31:09+00:00

Problem You find that the initial affine registration calculated by the intensity based registration software is failing. You would like to specify an alternative. Solution Calculate an alternative affine transformation and supply that to the registration software. Steps:

warping_manual:calculating_an_initial_surface-based_registration_using_amira

Anonymous (anonymous@undisclosed.example.com) — 2019-03-28T10:27:11+00:00

We quite often use the surface based registration option in Amira to generate an initial alignment for cases where CMTK's affine registration fails. * Take your starter image and smooth/downsample to e.g. 3 um voxels. * Then generate a surface – you can use the Isosurface viewer module

warping_manual:compile_cmtk_tools

Anonymous (anonymous@undisclosed.example.com) — 2016-02-01T12:58:34+00:00

CMTK can binaries can be downloaded from the project homepage and are generally the way to go. On a mac we currently use the dmg installer for macosx 10.6 NOT macports. However you may also want to compile from scratch. On a mac this looks like this: Prerequisites

warping_manual:custom_command_lines_for_munger.pl

Anonymous (anonymous@undisclosed.example.com) — 2013-12-10T11:59:45+00:00

When you use the simple RegistrationRunner.app gui it makes a subdirectory in your main registration directory called commands. Whenever you start a new job it saves a .command file in this directory. This file contains the actual commandline that is used to invoke the registration software. There are 3 layers:

warping_manual:finding_the_plane_of_symmetry

Anonymous (anonymous@undisclosed.example.com) — 2013-12-10T11:59:45+00:00

Torsten has a tool called symmetry_plane that can be used to find the single plane of symmetry separating the image. The command line options look like this: > symmetry_plane -h Compute volume image symmetry plane. (C) 1997-2007 Torsten Rohlfing [build 3866M] Usage: symmetry_plane [options] imagepath Supported options: -h, --help Print this help text. -v, --verbose Turn on verbosity mode. Optimization: -a, --accuracy Accuracy (final optimization step size in [mm]. -s, --sampling Resampl…

warping_manual:landmarks_registration_in_amira

Anonymous (anonymous@undisclosed.example.com) — 2013-12-10T11:59:45+00:00

Intro For simple visualisation reasons it may be helpful to align two brains by choosing corresponding points in both. This can be done in Amira using the Landmarks and LandmarksWarp objects. See Tutorial 2.7 Warping and Registration Using Landmarks in Amira's help.

warping_manual:landmarks_registration_with_cmtk

Anonymous (anonymous@undisclosed.example.com) — 2013-12-10T11:59:45+00:00

CMTK used to have a landmarks based registration tool based on VTK code called align_landmarks. In combination with munger.pl this allowed initialisation of CMTK affine/warp registrations by an initial affine calculated from pairs of landmarks in each sample brain and a corresponding set in a reference brain.

warping_manual:making_template_brains

Anonymous (anonymous@undisclosed.example.com) — 2013-12-10T11:59:45+00:00

In the past we have made template brains using the following general scheme * Choose 5 really good brains * ... and 46 good brains * register all 50 against each of the 5 seed brains * Score all the registrations * Pick the brain which had the most successes

warping_manual:plotting_3d_objects_in_r

Anonymous (anonymous@undisclosed.example.com) — 2014-01-06T20:49:34+00:00

RGL library R has a nice library for plotting in 3d called rgl (because it wraps the OpenGL framework). The functions are pretty self explanatory. There are 2 interfaces with separate sets of functions that are named either rgl.* or *3d. The *3d are slightly higher level and may be preferable. However they were added later and my code therefore uses the rgl.* functions (for example in the plotneuron3d routine)

warping_manual:registration_gui

Anonymous (anonymous@undisclosed.example.com) — 2018-06-09T16:05:52+00:00

Cross Platform GUI We now recommend a Registration GUI that is distributed as a plugin for Fiji (see below for details of the Classic MacOSX GUI). Currently this provides a simple way to install CMTK, select input images and registration parameters and construct appropriate registration scripts. However it does not actually run the final registration script - you still have to go to your operating system and run the script directly.

warping_manual:registration_parameters

Anonymous (anonymous@undisclosed.example.com) — 2013-12-10T11:59:45+00:00

There are number of things that are worth optimising when first starting to register a new kind of image. These include some of the parameters that are fed to the initial affine and warping registration. First it is important to make sure that the initial affine step is giving a good seed to the warp. If the affine isn't working well the warp may never be able to clean up the mess.

warping_manual:registration_quality

Anonymous (anonymous@undisclosed.example.com) — 2013-12-10T11:59:45+00:00

Marking the quality of a registration * Completely failed registration * Outline of the specimen correctly registered but major internal mismatch * Outline correct. Considerable internal mismatch * Outline correct. Minor internal mismatch - Potentially usable

warping_manual:start

Anonymous (anonymous@undisclosed.example.com) — 2018-06-21T12:51:28+00:00

Introduction In collaboration with Chris Potter (when he was in Liqun Luo's lab) and Torsten Rohlfing (formerly Stanford, SRI, now at Google) I have developed/applied techniques for mapping different fly brains into a common frame of reference and then using this for analysing the spatial relationships between different neurons. The software and approaches are likely to be of general relevance in neuroanatomy.

warping_manual:test_registration_data_set

Anonymous (anonymous@undisclosed.example.com) — 2013-12-10T11:59:45+00:00

The files at can be used as a sample to check that the registration software is working and to benchmark your machine. Testing Full Registration Use the GUI to run a full registration and reformat (command -awr 01). Make sure that you specify the location of the reference brain for this project when the