RFOVE

This repository is used to build a Docker image of the RFOVE tool described in: https://dx.doi.org/10.1016/j.imavis.2019.09.001

RFOVE is completely unsupervised, operates without any assumption or prior knowledge on the object’s shape and extends and improves the Decremental Ellipse Fitting Algorithm (DEFA). Both RFOVE and DEFA solve the multi-ellipse fitting problem by performing model selection that is guided by the minimization of the Akaike Information Criterion on a suitably defined shape complexity measure. However, in contrast to DEFA, RFOVE minimizes an objective function that allows for ellipses with higher degree of overlap and, thus, achieves better ellipse-based shape approximation.

The Docker image uses the original MATLAB implementation and makes it accessible without requiring any dependencies or license keys. LICENSE applies to the original implementation.

Docker instructions

Either build or pull the image:

• Build image: (only recommended for development)

  docker build --no-cache --tag kostrykin/rfove .

• Pull image: (recommended for production use)

  docker build --no-cache --tag kostrykin/rfove .

Run RFOVE:

docker run --rm -ti \
  -v /tmp/io:/io kostrykin/rfove \
  /rfove 250 0.1 0.2 201 /io/input.png /io/seg.tiff

In this example, the image /tmp/io/input.png is segmented and the results are written to /tmp/io/seg.tiff.