0.3.1

Added support for manually specifying the stitch edges in cases with four fragments, currently supported locations are ul (upper left), ur (upper right), ll (lower left) and lr (lower right). Simply specify these in the force_config.txt file as described in the readme. Also fixed a bug where the ranking of the solutions in the fragment configuration were not accurate.

0.3.0

Improved the tissue segmentation algorithm to filter out some adipose tissue for a better stitch between fragments. Additionally this release contains some minor bug fixes.

0.2.1

Fixed an important bug where theilsen edge detection did not work as expected when stitching two fragments.

0.2.0

Revamped the genetic algorithm by including more points along the stitching edge to guide the optimal transformation. PythoStitcher will now also automatically compute landmarks and provide information about the residual registration error after stitching.

0.1.1

Solved some minor bugs. This release will be used for the first version of the public container.

0.1.0

This release entails a significant refactoring of the final blending and reconstruction. This has roughly reduced computational overhead of these steps by half. In addition, users can now provide an input argument to specify at which resolution (in µm/pixel) they want to save the final reconstructed result.

0.0.1

First official pre-release of PythoStitcher. This should not expected to be stable.