Rigid registration algorithm for generating training/testing data for transfer function model
Project description
aics_tf_registration
Rigid registration algorithm for generating training/testing data for transfer function model
Features
- Rigid registration of
.tiff
confocal/fluorescent microscopy images, outputting cropped images containing their mutual field of view - Supports the following registration scenarios
- Images at the same or different resolutions and pixel dimensions/scales
- Full multichannel images based on a reference channel
- Specific channels in multichannel images based on seperate reference channel
- Multiple pairs of images with the same registration scenario at once
- Configuration of registration settings through easy-to-read
.yaml
file - Outputs composite of registered image for easy evaluation of results
Quick Start
In console (after installation):
run_alignment --config_path `path/to/config/file.yaml`
Installation
Stable Release: pip install aics_tf_registration
Development Head: pip install git+https://github.com/AllenCell/aics_tf_registration.git
Documentation
For full package documentation please visit AllenCell.github.io/aics_tf_registration.
Image Requirements for Registration
In order for the registration algorithm to produce accurate results, the images must have the following requirements:
- Images must be in
.tif
or.tiff
format. - The source and target images must be in separate folders and images that are to be registered to each other must share the same filename.
- Images must be 3D or 4D
- The field of view of either the source or target image must be wholly contained within the fov of the other (or cropped to be so with the settings in the config file)
- Rotation and mirroring of images (if necessary) to have matching orientations must either be done prior to registration or within the settings of the config file if it is consistent between different image pairs
- The resolution/voxel dimensions of the images, or at least the relative scaling differences between the source and target image, should be known to within approximately 3-4 decimal units
Free software: Allen Institute Software License
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
aics_tf_registration-0.1.1.tar.gz
(22.5 kB
view hashes)
Built Distribution
Close
Hashes for aics_tf_registration-0.1.1.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 514d32a579c50a482b14506109d05a33e9c5f57e68657befdc07f25b9336586f |
|
MD5 | 1d186cbf89cb121910e266b7835852a7 |
|
BLAKE2b-256 | 87874398c9d35b8bbc9fc3d9372f9331c547c9f91e30ca1cfbdd9c80aa520067 |
Close
Hashes for aics_tf_registration-0.1.1-py2.py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | d5226ca55fdade882ada250d480a545ac998313bc971ca0d0818d7933663cc23 |
|
MD5 | 4137d93589a8905bf640a15e40832f18 |
|
BLAKE2b-256 | a390ffca7eaf080db9cc76e8b527344f4acd5519449006310a55515b07437560 |