Introduction

Tools related to skeletonization of objects. The codes requires the ITK library to work and CMake to build.

Compilation

• Create a build directory for compilation

$ mkdir build

• Set the ITK_DIR variable inside the CMakeLists.txt. The exact path will depend on the location where ITK is installed on your machine. Run following command (with appropriate path/to/ITK) to set the variable.

$sed -i '10s/.*/set(ITK_DIR \/path\/to\/ITK)/' CMakeLists.txt

• Change into the build directory and generate make files using cmake

$ cd build
$ cmake -DCMAKE_BUILD_TYPE=Release ..

• Compile using make

$ make -j2

If all goes well you should now have skeltools executable insed the build folder.

Computing object centerline.

To compute (boundary distance weighted) medial curve for dinosaur sample image run

$ skeltool -priority distance -curve -input data/dinosaur.tif -smooth 0.2  -fillholes -outputFolder results -anchor aof

If everything runs without errors you should see dinosaur_distance_curve.tif inside results folder.

Sample dinosaur object in translucent blue with reddish colored medial curve inside

Computing object medial surface

To compute medial surface for dinosaur sample image run

$ skeltool -priority distance -surface -input data/dinosaur.tif -smooth 0.1  -fillholes -output results/dino_surface.tif -anchor aof

The output of the above command should be the medial surface dino_surface.tif inside the results folder.

Sample dinosaur object in translucent blue with reddish colored medial surface inside

Mapping object thickness to object boundary

MapToBoundaryImageFilter can be used to map features (for example object width) from medial surface to the object boundary. Example map-medial-thickness-to-boundary.cpp computes object width weighted medial skeleton of `dinosaur.tif' and maps it back to the boundary.

Visualization of object thickness to object boundary for dinosaur sample image

References

[1] Siddiqi, K., Bouix, S., Tannenbaum, A., & Zucker, S. W. (2002). Hamilton-jacobi skeletons. International Journal of Computer Vision, 48, 215-231.

[2] Dimitrov, P., Damon, J. N., & Siddiqi, K. (2003, June). Flux invariants for shape. In 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings. (Vol. 1, pp. I-I). IEEE.

[3] Siddiqi, K., & Pizer, S. (Eds.). (2008). Medial representations: mathematics, algorithms and applications (Vol. 37). Springer Science & Business Media.