Computer Vision News - November 2020

Registration Tool Comparison 5 as the Resampling, ROI, Crop Volume, Data, Mask Image, Subtract Images modules. Each of these can help adjust the data to improve the result of the registration. NiftyReg: the handy toolkit NiftyReg was created in 2009 as a command line tool for analysis and registration of medical images in nifti format. It’s available both with a CPU- and a GPU- (using CUDA) based implementation. Once the line tool is installed, it’s possible to access the source folder and type each of the commands in the list next to the argument -h to show the correspondent documentation. COMMAND USE reg_aladin Perform pairwise rigid or affine registration reg_f3d Perform pairwise non-linear registration reg_resample Resample a floating image in the space of a reference image given a transformation file reg_transform Convert transformation file from one type to another reg_jacobian Given a transformation file, generates Jacobian determinant or Jacobian matrix images. reg_average Compute average of images or affine matrices. reg_measure Compute similarity measure values between two input images. reg_tools Extra set of operations on images reg_gpuinfo Display information about the GPU on the current computer. reg_tools -h