Project

Efficient registration of large 3D images - lcreg

Goal: Rigid and affine multi-modality registration of large scalar 3D images is an import step for both medical and non-medical image processing. The distinguishing feature of this approach is its capability to efficiently register images that do not fit into system memory. lcreg is based on the optimisation of the local correlation similarity measure using a novel image encoding scheme fostering on-the-fly image compression and decompression. The software as well as a tutorial and test data can be obtained from https://pypi.org/project/lcreg/

Updates
0 new
2
Recommendations
0 new
0
Followers
0 new
8
Reads
0 new
598

Project log

Peter Rösch
added an update
The new release of the lcreg software adds support for Python 3.10 and systems based on Apple M1 chips. Furthermore, inofficial binary installers for bcolz are provided to facilitate lcreg installation.
Online-Ressources:
 
Peter Rösch
added an update
Version 0.1.3 of lcreg is available, see https://pypi.org/project/lcreg and https://github.com/p-roesch/lcreg. Apart from providing binaries for CPython version 3.8, adaptions for dependency updates have been performed.
 
Peter Rösch
added 2 project references
Peter Rösch
added a project goal
Rigid and affine multi-modality registration of large scalar 3D images is an import step for both medical and non-medical image processing. The distinguishing feature of this approach is its capability to efficiently register images that do not fit into system memory. lcreg is based on the optimisation of the local correlation similarity measure using a novel image encoding scheme fostering on-the-fly image compression and decompression. The software as well as a tutorial and test data can be obtained from https://pypi.org/project/lcreg/