A numerical, visual method for generating the(non-)linear separating isosurface of two classes of objects in two-dimensional space using Marching Squares and OpenCV

With regard to the separating isosurface of two classes of objects in two-dimensional space, the transition from rectilinear to curvilinear is documented.

Marching Squares and OpenCV were used to calculate the data that were used for illustrations in this paper.

It is found that the curvilinearity is adjustable via variable grid resolution, as well as via convolution such as Gaussian blur.

This paper's reference section is rather small, and the hunt for the literature is left to those who have access to the journals.