Skip to content

Registration of histological images using stitching and registration plugins in FIJI via PyImageJ

Notifications You must be signed in to change notification settings

uw-loci/automatic-histology-registration-pyimagej

Repository files navigation

Automatic Histological Image Registration

Registration of histological images using stitching and registration plugins in ImageJ/FIJI via PyImageJ.

Requirements

env.yml and PyImageJ and PyTorch if use CoMIR.

Generate intermediate modality for multimodal image registration

Registering images of different imaging modalities can be difficult. One can convert the images into a common intermediate modality (pseudo modality) using methods such as CoMIR.
To do this for histological images, patch extraction is often needed due to the big image size. Images are tiled into patches, converted into pseudo modality, stitched using Grid/Collection plugin in ImageJ/FIJI, and then registered. pseudo_modality.ipynb

Perform registration in pseudo modality

Once the images are converted into the common intermediate modality, conventional monomodal image registration such as SIFT feature matching can work well.
Monomdoal image registraion using SIFT landmark Correspondences in ImageJ/FIJI. ij_sift_registration.ipynb

About

Registration of histological images using stitching and registration plugins in FIJI via PyImageJ

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published