Zoltan Kato [    Research Group   |    Software   |    Publications   |    Teaching   |    Contact info   |    CV   |    Photography   ]

## Affine Shape Alignment Using Covariant Gaussian Densities

Please acknowledge the use of our programs by referring to the relevant publications.

This is the sample implementation and benchmark dataset of the binary image registration algorithm described in the following paper:

1. Csaba Domokos and Zoltan Kato. Affine Shape Alignment Using Covariant Gaussian Densities: A Direct Solution. Journal of Mathematical Imaging and Vision, 51(3) 385-399, March 2015. (Accepted)

The main code has been written by Csaba Domokos in Matlab. The code is released under the GNU General Public License.

The benchmark dataset has been created by Csaba Domokos. It contains 60 different template images containing compound shapes (15 images separately with 2, 3, 4 and 5 shapes) of size approx. 1000 X 1000 and 1435 observations generated synthetically by random affine transformations. Please cite the above publication whenever you use the dataset.

#### Usage Notes

The example.m matlab script shows an example how to use this demo implementation.

#### Benchmark dataset

This is a synthetic binary image dataset for testing affine registration methods. All images are in PNG format. There are 60 different template images around size 1000 x 1000. The observations were generated synthetically by applying randomly choosen affine transformations composed by

• 0, 10, ..., 350 degree of rotations;
• 0, 0.4, ..., 1.2 shearings;
• 0.5, 0.7, ..., 1.9 scalings;
• -20, 0, 20 translations along both axes.

The images are binary and the 0 and 1 represents the background and foreground, respectively!

Templates directory contains the template images in PNG file format. Naming convention: image????.png (image0001.png, image0002.png, ..., image0060.png)

Observations directory contains all observations for each template. Naming convention: observation????_????.png, where the first four digits are the template number, and last four digits count the number of the observation for the given template. Each observation has an associated data file with the same name and extension .dat.

 TemplateFileName filename of the corresponding template image ObservationFileName filename of the observation a11a12a13a21a22a23 the applied transformation matrix (row by row) Jacobian (number of pixels of the Observation) / (number of pixels of the Template)